1 This document describes *changes* to previous versions, that might
2 affect Exim's operation, with an unchanged configuration file. For new
3 options, and new features, see the NewStuff file next to this ChangeLog.
9 JH/01 Bug 1329: Fix format of Maildir-format filenames to match other mail-
10 related applications. Previously an "H" was used where available info
11 says that "M" should be, so change to match.
13 JH/02 Bug 2587: Fix pam expansion condition. Tainted values are commonly used
14 as arguments, so an implementation trying to copy these into a local
15 buffer was taking a taint-enforcement trap. Fix by using dynamically
16 created buffers. Similar fix for radius expansion condition.
18 JH/03 Bug 2586: Fix listcount expansion operator. Using tainted arguments is
19 reasonable, eg. to count headers. Fix by using dynamically created
20 buffers rather than a local. Do similar fixes for ACL actions "dcc",
21 "log_reject_target", "malware" and "spam"; the arguments are expanded
22 so could be handling tainted values.
24 JH/04 Bug 2590: Fix -bi (newaliases). A previous code rearrangement had
25 broken the (no-op) support for this sendmail command. Restore it
26 to doing nothing, silently, and returning good status.
28 JH/05 Bug 2593: Fix "vacation" in Exim filter. Previously, when a "once"
29 record path was given (or the default used) without a leading directory
30 path, an error occurred on trying to open it. Use the transport's working
33 JH/06 Bug 2594: Change the name used for certificate name checks in the smtp
34 transport. Previously it was the name on the DNS A-record; use instead
35 the head of the CNAME chain leading there (if there is one). This seems
36 to align better with RFC 6125.
38 JH/07 Bug 2597: Fix a resource leak. Using a lookup in obtaining a value for
39 smtp_accept_max_per_host allocated resources which were not released
40 when the limit was exceeded. This eventually crashed the daemon. Fix
41 by adding a relase action in that path.
43 JH/08 Bug 2598: Fix verify ACL condition. The options for the condition are
44 expanded; previously using tainted values was rejected. Fix by using
45 dynamically-created buffers.
47 JH/09 Relax restrictions on ACL verify condition needing access to message
48 headers. Previously they were only permitted in data and non-smtp ACLs;
49 permit also mime, dkim, prdr quit and notquit. Applies to header-syntax,
50 not_blind, header_sender and header_names_ascii verification.
52 JH/10 Bug 2603: Fix coding of string copying to only evaluate arguments once.
53 Previously a macro used one argument twice; when called with the
54 argument as an expression having side-effects, incorrect operation
55 resulted. Use an inlineable function.
57 JH/11 Bug 2604: Fix request to cutthrough-deliver when a connection is already
58 held open for a verify callout. Previously this wan not accounted for
59 and a corrupt onward SMTP conversation resulted.
61 JH/12 Bug 2607: Fix the ${srs_encode } expansion to handle quoted local_parts.
62 Previously they were embedded naively in the constructed address; when
63 needed, strip the quoting and quote the entire local_part.
64 Also make the inbound_srs expansion condition handle quoting.
66 JH/13 Fix dsearch "subdir" filter to ignore ".". Previously only ".." was
67 excluded, not matching the documentation.
69 JH/14 Bug 2606: Fix a segfault in sqlite lookups. When no, or a bad, filename
70 was given for the sqlite_dbfile a trap resulted.
72 JH/15 Bug 2620: Fix "spam" ACL condition. Previously, tainted values for the
73 "name" argument resulted in a trap. There is no reason to disallow such;
74 this was a coding error.
76 JH/16 Bug 2615: Fix pause during message reception, on systems that have been
77 suspended/resumed. The Linux CLOCK_MONOTONIC does not account for time
78 spent suspended, ignoring the Posix definition. Previously we assumed
79 it did and a constant offset from real time could be used as a correction.
80 Change to using the same clock source for the start-of-message and the
81 post-message next-tick-wait. Also change to using CLOCK_BOOTTIME if it
82 exists, just to get a clock slightly more aligned to reality.
84 JH/17 Bug 2295: Fix DKIM signing to always semicolon-terminate. Although the
85 RFC says it is optional some validators care. The missing char was not
86 intended but triggered by a line-wrap alignement. Discovery and fix by
87 Guillaume Outters, hacked on by JH.
89 JH/18 Bug 2617: Fix a taint trap in parse_fix_phrase(). Previously when the
90 name being quoted was tainted a trap would be taken. Fix by using
91 dynamicaly created buffers. The routine could have been called by a
92 rewrite with the "h" flag, by using the "-F" command-line option, or
93 by using a "name=" option on a control=submission ACL modifier.
95 JH/19 SPF: change the Authentication-Results expansion component to give
96 smtp.helo when the sender domain is empty. Previously it gave
99 JH/20 Bug 2631: ACL dnslist conditions now ignore and log any lookups returns
100 not in 127.0.0.0/8 to help in spotting list domains taken over by a
101 domain-parking registrar.
103 JH/21 Bug 2630: Fix eol-replacement string for the ${readsocket } expansion.
104 Previously when a whitespace character was specified it was not inserted
105 after removing the newline.
107 JH/22 Bug 2265: Force SNI usage for smtp transport DANE'd connections, to be
108 the domain part of the recipient address. This overrides any tls_sni
109 option set, which was previously used.
111 JH/23 Logging: with the +tls_sni log_selector, do not wrap the received SNI
114 JH/24 Bug 2634: Fix a taint trap seen on NetBSD: the testing coded for
115 is_tainted() had an off-by-one error in the overenthusiastic direction.
116 Find and fix by Gavan. Although NetBSD is not a supported platform for
117 4.94 this bug could affect other platforms.
119 PP/01 Fix default prime selection to be consistent.
120 One path used ike23 still, instead of exim.dev.20160529.3; now both
121 execution flows will use the same DH primes (currently
122 exim.dev.20160529.3).
124 JH/25 OpenSSL: Fix back-compatibility behaviour surrounding tls_certificates
125 option in smtp transport, to match the documentation. Previously
126 verification was not being done in some cases where it should have been.
128 JH/26 Bug 2646: fix a memory usage issue in ldap lookups. Previously, when more
129 than one server was defined and depending on the platform memory layout
130 details, an internal consistency trap could be hit while walking the list
133 JH/27 Bug 2648: fix the passing of an authenticator public-name through spool
134 files. The value is used by the authresults expansion item. Previously
135 if this was used in a router or transport, a crash could result.
137 JH/28 Fix spurious logging of select error. Some platforms, notably FreeBSD,
138 have a sufficient incidence of EINTR returns from select that an
139 interaction with other operations done by the main daemon loop exposed
140 a bug in the error-handling. This was benign apart from the log
143 JH/29 Bug 2675: add outgoing-interface I= element to deferred "==" log lines,
144 for consistency with delivered "=>" and failed "**" lines. While we're
145 there, handle PRX and TFO.
147 JH/30 Bug 2677: fix matching of long addresses. Since 4.93 a limit of 256 was
148 applied. This resulted, if any header-line rewrite rules were configured,
149 in a panic-log trigerrable by sending a message with a long address in
150 a header. Fix by increaing the arbitrary limit to larger than a single
151 (dewrapped) 5322 header line maximum size.
153 JH/31 The ESMTP option name advertised for the SUPPORT_EARLY_PIPE build option
154 is changed from X_PIPE_CONNECT to PIPE_CONNECT. This is in line with
155 RFC 6648 which deprecates X- options in protocols as a general practice.
156 Changeover between the implementations is handled by the mechanisms
159 JH/32 Bug 2599: fix delay of delivery to a local address where there is also
160 a remote which uses callout/hold. Previously the local was queued.
162 JH/33 Fix a taint trap in the ${listextract } expansion when the source data
165 JH/34 Fix the placement of a multiple-message delivery marker in the delivery
166 log line. The asterisk is now consistently appended to the remote IP
167 (and port, if given), and will also be provided on defer and fail log
168 lines. Previously it could be placed on the local IP if that was being
169 logged, and was only provided on delivery lines.
171 JH/35 Bug 2343: Harden exim_tidydb against corrupt wait- files.
173 JH/36 Bug 2687: Fix interpretation of multiple ^ chars in a plaintext
174 authenticator client_send option. Previously the next char, after a pair
175 was collapsed, was taken verbatim (so ^^^foo became ^^foo; ^^^^foo became
176 ^^\x00foo). Fixed to get ^\x00foo and ^^foo respectively to match the
177 documentation. There is still no way to get a leading ^ immediately
178 after a NUL (ie. for the password of a PLAIN method authenticator.
180 JH/37 Enforce the expected size, for fixed-size records read from hints-DB
181 files. For bad sizes read, delete the record and whine to paniclog.
183 JH/38 When logging an AUTH failure, as server, do not include sensitive
184 information. Previously, the credentials would be included if given
185 as part of the AUTH command line and an ACL denied authentidcation.
187 JH/39 Bug 2691: fix $local_part_data. When the matching list element
188 referred to a file, bad data was returned. This likely also affected
191 JH/40 The gsasl authenticator now supports caching of the salted password
192 generated by the client-side implementation. This required the addition
193 of a new variable: $auth4.
195 JH/41 Fix daemon SIGHUP on FreeBSD. Previously, a named socket for IPC was
196 left undeleted; the attempt to re-create it then failed - resulting in
197 the usual "SIGHUP tp have daemon reload configuration" to not work.
198 This affected any platform not supporting "abstract" Unix-domain
199 sockets (i.e. not Linux).
201 JH/42 Bug 2693: Harden against a peer which reneges on a 452 "too many
202 recipients" response to RCPT in a later response, with a 250. The
203 previous coding assumed this would not happen, and under PIPELINING
204 would result in both lost and duplicate recipients for a message.
206 JH/43 Bug 2694: Fix weighted distribution of work to multiple spamd servers.
207 Previously the weighting was incorrectly applied. Similar fix for socks
208 proxies. Found and fixed by Heiko Schlichting.
210 JH/44 Bug 2701: Fix list-expansion of dns_ipv4_lookup. Previously, it did
211 not handle sub-lists included using the +namedlist syntax. While
212 investigating, the same found for dns_trust_aa, dns_again_means_nonexist,
213 dnssec_require_domains, dnssec_request_domains, srv_fail_domains,
216 JH/45 Use a (new) separate store pool-pair for DKIM verify working data.
217 Previously the permanent pool was used, so the sore could not be freed.
218 This meant a connection with many messages would use continually-growing
225 JH/01 Avoid costly startup code when not strictly needed. This reduces time
226 for some exim process initialisations. It does mean that the logging
227 of TLS configuration problems is only done for the daemon startup.
229 JH/02 Early-pipelining support code is now included unless disabled in Makefile.
231 JH/03 DKIM verification defaults no long accept sha1 hashes, to conform to
232 RFC 8301. They can still be enabled, using the dkim_verify_hashes main
235 JH/04 Support CHUNKING from an smtp transport using a transport_filter, when
236 DKIM signing is being done. Previously a transport_filter would always
237 disable CHUNKING, falling back to traditional DATA.
239 JH/05 Regard command-line recipients as tainted.
241 JH/06 Bug 340: Remove the daemon pid file on exit, when due to SIGTERM.
243 JH/07 Bug 2489: Fix crash in the "pam" expansion condition. It seems that the
244 PAM library frees one of the arguments given to it, despite the
245 documentation. Therefore a plain malloc must be used.
247 JH/08 Bug 2491: Use tainted buffers for the transport smtp context. Previously
248 on-stack buffers were used, resulting in a taint trap when DSN information
249 copied from a received message was written into the buffer.
251 JH/09 Bug 2493: Harden ARC verify against Outlook, whick has been seen to mix
252 the ordering of its ARC headers. This caused a crash.
254 JH/10 Bug 2492: Use tainted memory for retry record when needed. Previously when
255 a new record was being constructed with information from the peer, a trap
258 JH/11 Bug 2494: Unset the default for dmarc_tld_file. Previously a naiive
259 installation would get error messages from DMARC verify, when it hit the
260 nonexistent file indicated by the default. Distros wanting DMARC enabled
261 should both provide the file and set the option.
262 Also enforce no DMARC verification for command-line sourced messages.
264 JH/12 Fix an uninitialised flag in early-pipelining. Previously connections
265 could, depending on the platform, hang at the STARTTLS response.
267 JH/13 Bug 2498: Reset a counter used for ARC verify before handling another
268 message on a connection. Previously if one message had ARC headers and
269 the following one did not, a crash could result when adding an
270 Authentication-Results: header.
272 JH/14 Bug 2500: Rewind some of the common-coding in string handling between the
273 Exim main code and Exim-related utities. The introduction of taint
274 tracking also did many adjustments to string handling. Since then, eximon
275 frequently terminated with an assert failure.
277 JH/15 When PIPELINING, synch after every hundred or so RCPT commands sent and
278 check for 452 responses. This slightly helps the inefficieny of doing
279 a large alias-expansion into a recipient-limited target. The max_rcpt
280 transport option still applies (and at the current default, will override
281 the new feature). The check is done for either cause of synch, and forces
282 a fast-retry of all 452'd recipients using a new MAIL FROM on the same
283 connection. The new facility is not tunable at this time.
285 JH/16 Fix the variables set by the gsasl authenticator. Previously a pointer to
286 library live data was being used, so the results became garbage. Make
287 copies while it is still usable.
289 JH/17 Logging: when the deliver_time selector ise set, include the DT= field
290 on delivery deferred (==) and failed (**) lines (if a delivery was
291 attemtped). Previously it was only on completion (=>) lines.
293 JH/18 Authentication: the gsasl driver not provides the $authN variables in time
294 for the expansion of the server_scram_iter and server_scram_salt options.
296 WB/01 SPF: DNS lookups for the obsolete SPF RR type done by the libspf2 library
297 are now specifically given a NO_DATA response without hitting the system
298 resolver. The library goes on to do the now-standard TXT lookup.
299 Use of dnsdb lookups is not affected.
301 JH/19 Bug 2507: Modules: on handling a dynamic-module (lookups) open failure,
302 only retrieve the errormessage once. Previously two calls to dlerror()
303 were used, and the second one (for mainlog/paniclog) retrieved null
306 JH/20 Taint checking: disallow use of tainted data for
307 - the appendfile transport file and directory options
308 - the pipe transport command
309 - the autoreply transport file, log and once options
310 - file names used by the redirect router (including filter files)
312 - paths used by single-key lookups
313 Previously this was permitted.
315 JH/21 Bug 2501: Fix init call in the heimdal authenticator. Previously it
316 adjusted the size of a major service buffer; this failed because the
317 buffer was in use at the time. Change to a compile-time increase in the
318 buffer size, when this authenticator is compiled into exim.
320 JH/22 Taint-checking: move to safe-mode taint checking on all platforms. The
321 previous fast-mode was untenable in the face of glibs using mmap to
322 support larger malloc requests.
324 PP/01 Update the openssl_options possible values through OpenSSL 1.1.1c.
325 New values supported, if defined on system where compiled:
326 allow_no_dhe_kex, cryptopro_tlsext_bug, enable_middlebox_compat,
327 no_anti_replay, no_encrypt_then_mac, prioritize_chacha, tlsext_padding
329 JH/23 Performance improvement in the initial phase of a two-pass queue run. By
330 running a limited number of proceses in parallel, a benefit is gained. The
331 amount varies with the platform hardware and load. The use of the option
332 queue_run_in_order means we cannot do this, as ordering becomes
335 JH/24 Bug 2524: fix the cyrus_sasl auth driver gssapi usage. A previous fix
336 had introduced a string-copy (for ensuring NUL-termination) which was not
337 appropriate for that case, which can include embedded NUL bytes in the
338 block of data. Investigation showed the copy to actually be needless, the
339 data being length-specified.
341 JH/25 Fix use of concurrent TLS connections under GnuTLS. When a callout was
342 done during a receiving connection, and both used TLS, global info was
343 used rather than per-connection info for tracking the state of data
344 queued for transmission. This could result in a connection hang.
346 JH/26 Fix use of the SIZE parameter on MAIL commands, on continued connections.
347 Previously, when delivering serveral messages down a single connection
348 only the first would provide a SIZE. This was due to the size information
349 not being properly tracked.
351 JH/27 Bug 2530: When operating in a timezone with sub-minute offset, such as
352 TAI (at 37 seconds currently), pretend to be in UTC for time-related
353 expansion and logging. Previously, spurious values such as a future
354 minute could be seen.
356 JH/28 Bug 2533: Fix expansion of ${tr } item. When called in some situations
357 it could crash from a null-deref. This could also affect the
358 ${addresses: } operator and ${readsock } item.
360 JH/29 Bug 2537: Fix $mime_part_count. When a single connection had a non-mime
361 message following a mime one, the variable was not reset.
363 JH/30 When an pipelined-connect fails at the first response, assume incorrect
364 cached capability (perhaps the peer reneged?) and immediately retry in
367 JH/31 Fix spurious detection of timeout while writing to transport filter.
369 JH/32 Bug 2541: Fix segfault on bad cmdline -f (sender) argument. Previously
370 an attempt to copy the string was made before checking it.
372 JH/33 Fix the dsearch lookup to return an untainted result. Previously the
373 taint of the lookup key was maintained; we now regard the presence in the
374 filesystem as sufficient validation.
376 JH/34 Fix the readsocket expansion to not segfault when an empty "options"
377 argument is supplied.
379 JH/35 The dsearch lookup now requires that the directory is an absolute path.
380 Previously this was not checked, and nonempty relative paths made an
381 access under Exim's current working directory.
383 JH/36 Bug 2554: Fix msg:defer event for the hosts_max_try_hardlimit case.
384 Previously no event was raised.
386 JH/37 Bug 2552: Fix the check on spool space during reception to use the SIZE
387 parameter supplied by the sender MAIL FROM command. Previously it was
388 ignored, and only the check_spool_space option value for the required
391 JH/38 Fix $dkim_key_length. This should, after a DKIM verification, present
392 the size of the signing public-key. Previously it was instead giving
393 the size of the signature hash.
395 JH/39 DKIM verification: the RFC 8301 restriction on sizes of RSA keys is now
396 the default. See the (new) dkim_verify_min_keysizes option.
398 JH/40 Fix a memory-handling bug: when a connection carried multiple messages
399 and an ACL use a lookup for checking either the local_part or domain,
400 stale data could be accessed. Ensure that variable references are
401 dropped between messages.
403 JH/41 Bug 2571: Fix SPA authenticator. Running as a server, an offset supplied
404 by the client was not checked as pointing within response data before
405 being used. A malicious client could thus cause an out-of-bounds read and
406 possibly gain authentication. Fix by adding the check.
408 JH/42 Internationalisation: change the default for downconversion in the smtp
409 transport to be "if needed". Previously it was "as previously set" for
410 the message, which usually meant "if needed" for message-submission but
411 "no" for everything else. However, MTAs have been seen using SMTPUTF8
412 even when the envelope addresses did not need it, resulting in forwarding
413 failures to non-supporting MTAs. A downconvert in such cases will be
414 a no-op on the addresses, merely dropping the use of SMTPUTF8 by the
415 transport. The change does mean that addresses needing conversion will
416 be converted when previously a delivery failure would occur.
418 JH/43 Fix possible long line in DSN. Previously when a very long SMTP error
419 response was received it would be used unchecked in a fail-DSN, violating
420 standards on line-length limits. Truncate if needed.
422 HS/01 Remove parameters of the link to www.open-spf.org. The linked form
423 doesn't work. (Additionally add a new main config option to configure the
430 JH/01 OpenSSL: With debug enabled output keying information sufficient, server
431 side, to decode a TLS 1.3 packet capture.
433 JH/02 OpenSSL: Suppress the sending of (stateful) TLS1.3 session tickets.
434 Previously the default library behaviour applied, sending two, each in
437 JH/03 Debug output for ACL now gives the config file name and line number for
440 JH/04 The default received_header_text now uses the RFC 8314 tls cipher clause.
442 JH/05 DKIM: ensure that dkim_domain elements are lowercased before use.
444 JH/06 Fix buggy handling of autoreply bounce_return_size_limit, and a possible
445 buffer overrun for (non-chunking) other transports.
447 JH/07 GnuTLS: Our use of late (post-handshake) certificate verification, under
448 TLS1.3, means that a server rejecting a client certificate is not visible
449 to the client until the first read of encrypted data (typically the
450 response to EHLO). Add detection for that case and treat it as a failed
451 TLS connection attempt, so that the normal retry-in-clear can work (if
452 suitably configured).
454 JB/01 Bug 2375: fix expansions of 822 addresses having comments in local-part
455 and/or domain. Found and fixed by Jason Betts.
457 JH/08 Add hardening against SRV & TLSA lookups the hit CNAMEs (a nonvalid
458 configuration). If a CNAME target was not a wellformed name pattern, a
461 JH/09 Logging: Fix initial listening-on line for multiple ports for an IP when
462 the OS reports them interleaved with other addresses.
464 JH/10 OpenSSL: Fix aggregation of messages. Previously, when PIPELINING was
465 used both for input and for a verify callout, both encrypted, SMTP
466 responses being sent by the server could be lost. This resulted in
467 dropped connections and sometimes bounces generated by a peer sending
470 JH/11 Harden plaintext authenticator against a badly misconfigured client-send
471 string. Previously it was possible to cause undefined behaviour in a
472 library routine (usually a crash). Found by "zerons".
474 JH/12 Bug 2384: fix "-bP smtp_receive_timeout". Previously it returned no
477 JH/13 Bug 2386: Fix builds with Dane under LibreSSL 2.9.0 onward. Some old
478 API was removed, so update to use the newer ones.
480 JH/14 Bug 1891: Close the log file if receiving a non-smtp message, without
481 any timeout set, is taking a long time. Previously we would hang on to a
482 rotated logfile "forever" if the input was arriving with long gaps
483 (a previous attempt to fix addressed lack, for a long time, of initial
486 HS/01 Bug 2390: Use message_id for tempfile creation to avoid races in a
487 shared (NFS) environment. The length of the tempfile name is now
488 4 + 16 ("hdr.$message_exim_id") which might break on file
489 systems which restrict the file name length to lower values.
492 HS/02 Bug 2390: Use message_id for tempfile creation to avoid races in a
493 shared (NFS) environment.
495 HS/03 Bug 2392: exigrep does case sensitive *option* processing (as it
496 did for all versions <4.90). Notably -M, -m, --invert, -I may be
499 JH/15 Use unsigned when creating bitmasks in macros, to avoid build errors
500 on some platforms for bit 31.
502 JH/16 GnuTLS: rework ciphersuite strings under recent library versions. Thanks
503 to changes apparently associated with TLS1.3 handling some of the APIs
504 previously used were either nonfunctional or inappropriate. Strings
505 like TLS1.3:ECDHE_SECP256R1__RSA_PSS_RSAE_SHA256__AES_256_GCM__AEAD:256
506 and TLS1.2:ECDHE_SECP256R1__RSA_SHA256__AES_128_CBC__SHA256:128 replace
507 the previous TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256 .
508 This affects log line X= elements, the $tls_{in,out}_cipher variables,
509 and the use of specific cipher names in the encrypted= ACL condition.
511 JH/17 OpenSSL: the default openssl_options now disables ssl_v3.
513 JH/18 GnuTLS: fix $tls_out_ocsp under hosts_request_ocsp. Previously the
514 verification result was not updated unless hosts_require_ocsp applied.
516 JH/19 Bug 2398: fix listing of a named-queue. Previously, even with the option
517 queue_list_requires_admin set to false, non-admin users were denied the
520 JH/20 Bug 2389: fix server advertising of usable certificates, under GnuTLS in
521 directory-of-certs mode. Previously they were advertised despite the
524 JH/21 The smtp transport option "hosts_noproxy_tls" is now unset by default.
525 A single TCP connection by a client will now hold a TLS connection open
526 for multiple message deliveries, by default. Previously the default was to
529 JH/22 The smtp transport option "hosts_try_dane" now enables all hosts by
530 default. If built with the facility, DANE will be used. The facility
531 SUPPORT_DANE is now enabled in the prototype build Makefile "EDITME".
533 JH/23 The build default is now for TLS to be included; the SUPPORT_TLS define
534 is replaced with DISABLE_TLS. Either USE_GNUTLS or (the new) USE_OPENSSL
535 must be defined and you must still, unless you define DISABLE_TLS, manage
536 the the include-dir and library-file requirements that go with that
537 choice. Non-TLS builds are still supported.
539 JH/24 Fix duplicated logging of peer name/address, on a transport connection-
542 JH/25 The smtp transport option "hosts_try_fastopen" now enables all hosts by
543 default. If the platform supports and has the facility enabled, it will
544 be requested on all coneections.
546 JH/26 The PIPE_CONNECT facility is promoted from experimental status and is now
547 controlled by the build-time option SUPPORT_PIPE_CONNECT.
549 PP/01 Unbreak heimdal_gssapi, broken in 4.92.
551 JH/27 Bug 2404: Use the main-section configuration option "dsn_from" for
552 success-DSN messages. Previously the From: header was always the default
553 one for these; the option was ignored.
555 JH/28 Fix the timeout on smtp response to apply to the whole response.
556 Previously it was reset for every read, so a teergrubing peer sending
557 single bytes within the time limit could extend the connection for a
558 long time. Credit to Qualsys Security Advisory Team for the discovery.
560 JH/29 Fix DSN Final-Recipient: field. Previously it was the post-routing
561 delivery address, which leaked information of the results of local
562 forwarding. Change to the original envelope recipient address, per
565 JH/30 Bug 2411: Fix DSN generation when RFC 3461 failure notification is
566 requested. Previously not bounce was generated and a log entry of
567 error ignored was made.
569 JH/31 Avoid re-expansion in ${sort } expansion. (CVE-2019-13917)
571 JH/32 Introduce a general tainting mechanism for values read from the input
572 channel, and values derived from them. Refuse to expand any tainted
573 values, to catch one form of exploit.
575 JH/33 Bug 2413: Fix dkim_strict option. Previously the expansion result
576 was unused and the unexpanded text used for the test. Found and
577 fixed by Ruben Jenster.
579 JH/34 Fix crash after TLS shutdown. When the TCP/SMTP channel was left open,
580 an attempt to use a TLS library read routine dereffed a nul pointer,
583 JH/35 Bug 2409: filter out-of-spec chars from callout response before using
584 them in our smtp response.
586 JH/36 Have the general router option retry_use_local_part default to true when
587 any of the restrictive preconditions are set (to anything). Previously it
588 was only for check_local user. The change removes one item of manual
589 configuration which is required for proper retries when a remote router
590 handles a subset of addresses for a domain.
592 JH/37 Appendfile: when evaluating quota use (non-quota_size_regex) take the file
593 link count into consideration.
595 HS/04 Fix handling of very log lines in -H files. If a -<key> <value> line
596 caused the extension of big_buffer, the following lines were ignored.
598 JH/38 Bug 1395: Teach the DNS negative-cache about TTL value from the SOA in
599 accordance with RFC 2308. Previously there was no expiry, so a longlived
600 receive process (eg. due to ACL delays) versus a short SOA value could
603 HS/05 Handle trailing backslash gracefully. (CVE-2019-15846)
605 JH/39 Promote DMARC support to mainline.
607 JH/40 Bug 2452: Add a References: header to DSNs.
609 JH/41 With GnuTLS 3.6.0 (and later) do not attempt to manage Diffie-Hellman
610 parameters. The relevant library call is documented as "Deprecated: This
611 function is unnecessary and discouraged on GnuTLS 3.6.0 or later. Since
612 3.6.0, DH parameters are negotiated following RFC7919."
614 HS/06 Change the default of dnssec_request_domains to "*"
616 JH/42 Bug 2545: Fix CHUNKING for all RCPT commands rejected. Previously we
617 carried on and emitted a BDAT command, even when PIPELINING was not
620 JH/43 Bug 2465: Fix taint-handling in dsearch lookup. Previously a nontainted
621 buffer was used for the filename, resulting in a trap when tainted
622 arguments (eg. $domain) were used.
624 JH/44 With OpenSSL 1.1.1 (onwards) disable renegotiation for TLS1.2 and below;
625 recommended to avoid a possible server-load attack. The feature can be
626 re-enabled via the openssl_options main cofiguration option.
628 JH/45 local_scan API: documented the current smtp_printf() call. This changed
629 for version 4.90 - adding a "more data" boolean to the arguments.
630 Bumped the ABI version number also, this having been missed previously;
631 release versions 4.90 to 4.92.3 inclusive were effectively broken in
632 respect of usage of smtp_printf() by either local_scan code or libraries
633 accessed via the ${dlfunc } expansion item. Both will need coding
634 adjustment for any calls to smtp_printf() to match the new function
635 signature; a FALSE value for the new argument is always safe.
637 JH/46 FreeBSD: fix use of the sendfile() syscall. The shim was not updating
638 the file-offset (which the Linux syscall does, and exim expects); this
639 resulted in an indefinite loop.
641 JH/47 ARC: fix crash in signing, triggered when a configuration error failed
642 to do ARC verification. The Authentication-Results: header line added
643 by the configuration then had no ARC item.
649 JH/01 Remove code calling the customisable local_scan function, unless a new
650 definition "HAVE_LOCAL_SCAN=yes" is present in the Local/Makefile.
652 JH/02 Bug 1007: Avoid doing logging from signal-handlers, as that can result in
653 non-signal-safe functions being used.
655 JH/03 Bug 2269: When presented with a received message having a stupidly large
656 number of DKIM-Signature headers, disable DKIM verification to avoid
657 a resource-consumption attack. The limit is set at twenty.
659 JH/04 Add variables $arc_domains, $arc_oldest_pass for ARC verify. Fix the
660 report of oldest_pass in ${authres } in consequence, and separate out
661 some descriptions of reasons for verification fail.
663 JH/05 Bug 2273: Cutthrough delivery left a window where the received messsage
664 files in the spool were present and unlocked. A queue-runner could spot
665 them, resulting in a duplicate delivery. Fix that by doing the unlock
666 after the unlink. Investigation by Tim Stewart. Take the opportunity to
667 add more error-checking on spoolfile handling while that code is being
670 PP/01 Refuse to open a spool data file (*-D) if it's a symlink.
671 No known attacks, no CVE, this is defensive hardening.
673 JH/06 Bug 2275: The MIME ACL unlocked the received message files early, and
674 a queue-runner could start a delivery while other operations were ongoing.
675 Cutthrough delivery was a common victim, resulting in duplicate delivery.
676 Found and investigated by Tim Stewart. Fix by using the open message data
677 file handle rather than opening another, and not locally closing it (which
678 releases a lock) for that case, while creating the temporary .eml format
679 file for the MIME ACL. Also applies to "regex" and "spam" ACL conditions.
681 JH/07 Bug 177: Make a random-recipient callout success visible in ACL, by setting
682 $sender_verify_failure/$recipient_verify_failure to "random".
684 JH/08 When generating a selfsigned cert, use serial number 1 since zero is not
687 JH/09 Bug 2274: Fix logging of cmdline args when starting in an unlinked cwd.
688 Previously this would segfault.
690 JH/10 Fix ARC signing for case when DKIM signing failed. Previously this would
693 JH/11 Bug 2264: Exim now only follows CNAME chains one step by default. We'd
694 like zero, since the resolver should be doing this for us, But we need one
695 as a CNAME but no MX presence gets the CNAME returned; we need to check
696 that doesn't point to an MX to declare it "no MX returned" rather than
697 "error, loop". A new main option is added so the older capability of
698 following some limited number of chain links is maintained.
700 JH/12 Add client-ip info to non-pass iprev ${authres } lines.
702 JH/13 For receent Openssl versions (1.1 onward) use modern generic protocol
703 methods. These should support TLS 1.3; they arrived with TLS 1.3 and the
704 now-deprecated earlier definitions used only specified the range up to TLS
705 1.2 (in the older-version library docs).
707 JH/14 Bug 2284: Fix DKIM signing for body lines starting with a pair of dots.
709 JH/15 Rework TLS client-side context management. Stop using a global, and
710 explicitly pass a context around. This enables future use of TLS for
711 connections to service-daemons (eg. malware scanning) while a client smtp
712 connection is using TLS; with cutthrough connections this is quite likely.
714 JH/16 Fix ARC verification to do AS checks in reverse order.
716 JH/17 Support a "tls" option on the ${readsocket } expansion item.
718 JH/18 Bug 2287: Fix the protocol name (eg utf8esmtp) for multiple messages
719 using the SMTPUTF8 option on their MAIL FROM commands, in one connection.
720 Previously the "utf8" would be re-prepended for every additional message.
722 JH/19 Reject MAIL FROM commands with SMTPUTF8 when the facility was not advertised.
723 Previously thery were accepted, resulting in issues when attempting to
724 forward messages to a non-supporting MTA.
726 PP/02 Let -n work with printing macros too, not just options.
728 JH/20 Bug 2296: Fix cutthrough for >1 address redirection. Previously only
729 one parent address was copied, and bogus data was used at delivery-logging
730 time. Either a crash (after delivery) or bogus log data could result.
731 Discovery and analysis by Tim Stewart.
733 PP/03 Make ${utf8clean:} expansion operator detect incomplete final character.
734 Previously if the string ended mid-character, we did not insert the
735 promised '?' replacement.
737 PP/04 Documentation: current string operators work on bytes, not codepoints.
739 JH/21 Change as many as possible of the global flags into one-bit bitfields; these
740 should pack well giving a smaller memory footprint so better caching and
741 therefore performance. Group the declarations where this can't be done so
742 that the byte-sized flag variables are not interspersed among pointer
743 variables, giving a better chance of good packing by the compiler.
745 JH/22 Bug 1896: Fix the envelope from for DMARC forensic reports to be possibly
746 non-null, to avoid issues with sites running BATV. Previously reports were
747 sent with an empty envelope sender so looked like bounces.
749 JH/23 Bug 2318: Fix the noerror command within filters. It wasn't working.
750 The ignore_error flag wasn't being returned from the filter subprocess so
751 was not set for later routers. Investigation and fix by Matthias Kurz.
753 JH/24 Bug 2310: Raise a msg:fail:internal event for each undelivered recipient,
754 and a msg:complete for the whole, when a message is manually removed using
755 -Mrm. Developement by Matthias Kurz, hacked on by JH.
757 JH/25 Avoid fixed-size buffers for pathnames in DB access. This required using
758 a "Gnu special" function, asprintf() in the DB utility binary builds; I
759 hope that is portable enough.
761 JH/26 Bug 2311: Fix DANE-TA verification under GnuTLS. Previously it was also
762 requiring a known-CA anchor certificate; make it now rely entirely on the
763 TLSA as an anchor. Checking the name on the leaf cert against the name
764 on the A-record for the host is still done for TA (but not for EE mode).
766 JH/27 Fix logging of proxy address. Previously, a pointless "PRX=[]:0" would be
767 included in delivery lines for non-proxied connections, when compiled with
768 SUPPORT_SOCKS and running with proxy logging enabled.
770 JH/28 Bug 2314: Fire msg:fail:delivery event even when error is being ignored.
771 Developement by Matthias Kurz, tweaked by JH. While in that bit of code,
772 move the existing event to fire before the normal logging of message
773 failure so that custom logging is bracketed by normal logging.
775 JH/29 Bug 2322: A "fail" command in a non-system filter (file) now fires the
776 msg:fail:internal event. Developement by Matthias Kurz.
778 JH/30 Bug 2329: Increase buffer size used for dns lookup from 2k, which was
779 far too small for todays use of crypto signatures stored there. Go all
780 the way to the max DNS message size of 64kB, even though this might be
781 overmuch for IOT constrained device use.
783 JH/31 Fix a bad use of a copy function, which could be used to pointlessly
784 copy a string over itself. The library routine is documented as not
785 supporting overlapping copies, and on MacOS it actually raised a SIGABRT.
787 JH/32 For main options check_spool_space and check_inode_space, where the
788 platform supports 64b integers, support more than the previous 2^31 kB
789 (i.e. more than 2 TB). Accept E, P and T multipliers in addition to
790 the previous G, M, k.
792 JH/33 Bug 2338: Fix the cyrus-sasl authenticator to fill in the
793 $authenticated_fail_id variable on authentication failure. Previously
796 JH/34 Increase RSA keysize of autogen selfsign cert from 1024 to 2048. RHEL 8.0
797 OpenSSL didn't want to use such a weak key. Do for GnuTLS also, and for
798 more-modern GnuTLS move from GNUTLS_SEC_PARAM_LOW to
799 GNUTLS_SEC_PARAM_MEDIUM.
801 JH/35 OpenSSL: fail the handshake when SNI processing hits a problem, server
802 side. Previously we would continue as if no SNI had been received.
804 JH/36 Harden the handling of string-lists. When a list consisted of a sole
805 "<" character, which should be a list-separator specification, we walked
806 off past the nul-terimation.
808 JH/37 Bug 2341: Send "message delayed" warning MDNs (restricted to external
809 causes) even when the retry time is not yet met. Previously they were
810 not, meaning that when (say) an account was over-quota and temp-rejecting,
811 and multiple senders' messages were queued, only one sender would get
812 notified on each configured delay_warning cycle.
814 JH/38 Bug 2351: Log failures to extract envelope addresses from message headers.
816 JH/39 OpenSSL: clear the error stack after an SSL_accept(). With anon-auth
817 cipher-suites, an error can be left on the stack even for a succeeding
818 accept; this results in impossible error messages when a later operation
821 AM/01 Bug 2359: GnuTLS: repeat lowlevel read and write operations while they
822 return error codes indicating retry. Under TLS1.3 this becomes required.
824 JH/40 Fix the feature-cache refresh for EXPERIMENTAL_PIPE_CONNECT. Previously
825 it only wrote the new authenticators, resulting in a lack of tracking of
826 peer changes of ESMTP extensions until the next cache flush.
828 JH/41 Fix the loop reading a message header line to check for integer overflow,
829 and more-often against header_maxsize. Previously a crafted message could
830 induce a crash of the recive process; now the message is cleanly rejected.
832 JH/42 Bug 2366: Fix the behaviour of the dkim_verify_signers option. It had
833 been totally disabled for all of 4.91. Discovery and fix by "Mad Alex".
839 GF/01 DEFER rather than ERROR on redis cluster MOVED response.
840 When redis_servers is set to a list of > 1 element, and the Redis servers
841 in that list are in cluster configuration, convert the REDIS_REPLY_ERROR
842 case of MOVED into a DEFER case instead, thus moving the query onto the
843 next server in the list. For a cluster of N elements, all N servers must
844 be defined in redis_servers.
846 GF/02 Catch and remove uninitialized value warning in exiqsumm
847 Check for existence of @ARGV before looking at $ARGV[0]
849 JH/01 Replace the store_release() internal interface with store_newblock(),
850 which internalises the check required to safely use the old one, plus
851 the allocate and data copy operations duplicated in both (!) of the
852 extant use locations.
854 JH/02 Disallow '/' characters in queue names specified for the "queue=" ACL
855 modifier. This matches the restriction on the commandline.
857 JH/03 Fix pgsql lookup for multiple result-tuples with a single column.
858 Previously only the last row was returned.
860 JH/04 Bug 2217: Tighten up the parsing of DKIM signature headers. Previously
861 we assumed that tags in the header were well-formed, and parsed the
862 element content after inspecting only the first char of the tag.
863 Assumptions at that stage could crash the receive process on malformed
866 JH/05 Bug 2215: Fix crash associated with dnsdb lookup done from DKIM ACL.
867 While running the DKIM ACL we operate on the Permanent memory pool so that
868 variables created with "set" persist to the DATA ACL. Also (at any time)
869 DNS lookups that fail create cache records using the Permanent pool. But
870 expansions release any allocations made on the current pool - so a dnsdb
871 lookup expansion done in the DKIM ACL releases the memory used for the
872 DNS negative-cache, and bad things result. Solution is to switch to the
873 Main pool for expansions.
874 While we're in that code, add checks on the DNS cache during store_reset,
875 active in the testsuite.
876 Problem spotted, and debugging aided, by Wolfgang Breyha.
878 JH/06 Fix issue with continued-connections when the DNS shifts unreliably.
879 When none of the hosts presented to a transport match an already-open
880 connection, close it and proceed with the list. Previously we would
881 queue the message. Spotted by Lena with Yahoo, probably involving
884 JH/07 Bug 2214: Fix SMTP responses resulting from non-accept result of MIME ACL.
885 Previously a spurious "250 OK id=" response was appended to the proper
888 JH/08 The "support for" informational output now, which built with Content
889 Scanning support, has a line for the malware scanner interfaces compiled
890 in. Interface can be individually included or not at build time.
892 JH/09 The "aveserver", "kavdaemon" and "mksd" interfaces are now not included
893 by the template makefile "src/EDITME". The "STREAM" support for an older
894 ClamAV interface method is removed.
896 JH/10 Bug 2223: Fix mysql lookup returns for the no-data case (when the number of
897 rows affected is given instead).
899 JH/11 The runtime Berkeley DB library version is now additionally output by
900 "exim -d -bV". Previously only the compile-time version was shown.
902 JH/12 Bug 2230: Fix cutthrough routing for nonfirst messages in an initiating
903 SMTP connection. Previously, when one had more recipients than the
904 first, an abortive onward connection was made. Move to full support for
905 multiple onward connections in sequence, handling cutthrough connection
906 for all multi-message initiating connections.
908 JH/13 Bug 2229: Fix cutthrough routing for nonstandard port numbers defined by
909 routers. Previously, a multi-recipient message would fail to match the
910 onward-connection opened for the first recipient, and cause its closure.
912 JH/14 Bug 2174: A timeout on connect for a callout was also erroneously seen as
913 a timeout on read on a GnuTLS initiating connection, resulting in the
914 initiating connection being dropped. This mattered most when the callout
915 was marked defer_ok. Fix to keep the two timeout-detection methods
918 JH/15 Relax results from ACL control request to enable cutthrough, in
919 unsupported situations, from error to silently (except under debug)
920 ignoring. This covers use with PRDR, frozen messages, queue-only and
923 HS/01 Fix Buffer overflow in base64d() (CVE-2018-6789)
925 JH/16 Fix bug in DKIM verify: a buffer overflow could corrupt the malloc
926 metadata, resulting in a crash in free().
928 PP/01 Fix broken Heimdal GSSAPI authenticator integration.
929 Broken in f2ed27cf5, missing an equals sign for specified-initialisers.
930 Broken also in d185889f4, with init system revamp.
932 JH/17 Bug 2113: Fix conversation closedown with the Avast malware scanner.
933 Previously we abruptly closed the connection after reading a malware-
934 found indication; now we go on to read the "scan ok" response line,
937 JH/18 Bug 2239: Enforce non-usability of control=utf8_downconvert in the mail
938 ACL. Previously, a crash would result.
940 JH/19 Speed up macro lookups during configuration file read, by skipping non-
941 macro text after a replacement (previously it was only once per line) and
942 by skipping builtin macros when searching for an uppercase lead character.
944 JH/20 DANE support moved from Experimental to mainline. The Makefile control
945 for the build is renamed.
947 JH/21 Fix memory leak during multi-message connections using STARTTLS. A buffer
948 was allocated for every new TLS startup, meaning one per message. Fix
949 by only allocating once (OpenSSL) or freeing on TLS-close (GnuTLS).
951 JH/22 Bug 2236: When a DKIM verification result is overridden by ACL, DMARC
952 reported the original. Fix to report (as far as possible) the ACL
953 result replacing the original.
955 JH/23 Fix memory leak during multi-message connections using STARTTLS under
956 OpenSSL. Certificate information is loaded for every new TLS startup,
957 and the resources needed to be freed.
959 JH/24 Bug 2242: Fix exim_dbmbuild to permit directoryless filenames.
961 JH/25 Fix utf8_downconvert propagation through a redirect router. Previously it
964 JH/26 Bug 2253: For logging delivery lines under PRDR, append the overall
965 DATA response info to the (existing) per-recipient response info for
966 the "C=" log element. It can have useful tracking info from the
967 destination system. Patch from Simon Arlott.
969 JH/27 Bug 2251: Fix ldap lookups that return a single attribute having zero-
970 length value. Previously this would segfault.
972 HS/02 Support Avast multiline protoocol, this allows passing flags to
973 newer versions of the scanner.
975 JH/28 Ensure that variables possibly set during message acceptance are marked
976 dead before release of memory in the daemon loop. This stops complaints
977 about them when the debug_store option is enabled. Discovered specifically
978 for sender_rate_period, but applies to a whole set of variables.
979 Do the same for the queue-runner and queue-list loops, for variables set
980 from spool message files. Do the same for the SMTP per-message loop, for
981 certain variables indirectly set in ACL operations.
983 JH/29 Bug 2250: Fix a longstanding bug in heavily-pipelined SMTP input (such
984 as a multi-recipient message from a mailinglist manager). The coding had
985 an arbitrary cutoff number of characters while checking for more input;
986 enforced by writing a NUL into the buffer. This corrupted long / fast
987 input. The problem was exposed more widely when more pipelineing of SMTP
988 responses was introduced, and one Exim system was feeding another.
989 The symptom is log complaints of SMTP syntax error (NUL chars) on the
990 receiving system, and refused recipients seen by the sending system
991 (propating to people being dropped from mailing lists).
992 Discovered and pinpointed by David Carter.
994 JH/30 The (EXPERIMENTAL_DMARC) variable $dmarc_ar_header is withdrawn, being
995 replaced by the ${authresults } expansion.
997 JH/31 Bug 2257: Fix pipe transport to not use a socket-only syscall.
999 HS/03 Set a handler for SIGTERM and call exit(3) if running as PID 1. This
1000 allows proper process termination in container environments.
1002 JH/32 Bug 2258: Fix spool_wireformat in combination with LMTP transport.
1003 Previously the "final dot" had a newline after it; ensure it is CR,LF.
1005 JH/33 SPF: remove support for the "spf" ACL condition outcome values "err_temp"
1006 and "err_perm", deprecated since 4.83 when the RFC-defined words
1007 "temperror" and "permerror" were introduced.
1009 JH/34 Re-introduce enforcement of no cutthrough delivery on transports having
1010 transport-filters or DKIM-signing. The restriction was lost in the
1011 consolidation of verify-callout and delivery SMTP handling.
1012 Extend the restriction to also cover ARC-signing.
1014 JH/35 Cutthrough: for a final-dot response timeout (and nonunderstood responses)
1015 in defer=pass mode supply a 450 to the initiator. Previously the message
1018 PP/02 DANE: add dane_require_tls_ciphers SMTP Transport option; if unset,
1019 tls_require_ciphers is used as before.
1021 HS/03 Malware Avast: Better match the Avast multiline protocol. Add
1022 "pass_unscanned". Only tmpfails from the scanner are written to
1023 the paniclog, as they may require admin intervention (permission
1024 denied, license issues). Other scanner errors (like decompression
1025 bombs) do not cause a paniclog entry.
1027 JH/36 Fix reinitialisation of DKIM logging variable between messages.
1028 Previously it was possible to log spurious information in receive log
1031 JH/37 Bug 2255: Revert the disable of the OpenSSL session caching. This
1032 triggered odd behaviour from Outlook Express clients.
1034 PP/03 Add util/renew-opendmarc-tlds.sh script for safe renewal of public
1037 JH/38 DKIM: accept Ed25519 pubkeys in SubjectPublicKeyInfo-wrapped form,
1038 since the IETF WG has not yet settled on that versus the original
1039 "bare" representation.
1041 JH/39 Fix syslog logging for syslog_timestamp=no and log_selector +millisec.
1042 Previously the millisecond value corrupted the output.
1043 Fix also for syslog_pid=no and log_selector +pid, for which the pid
1044 corrupted the output.
1050 JH/01 Rework error string handling in TLS interface so that the caller in
1051 more cases is responsible for logging. This permits library-sourced
1052 string to be attached to addresses during delivery, and collapses
1053 pairs of long lines into single ones.
1055 PP/01 Allow PKG_CONFIG_PATH to be set in Local/Makefile and use it correctly
1056 during configuration. Wildcards are allowed and expanded.
1058 JH/02 Rework error string handling in DKIM to pass more info back to callers.
1059 This permits better logging.
1061 JH/03 Rework the transport continued-connection mechanism: when TLS is active,
1062 do not close it down and have the child transport start it up again on
1063 the passed-on TCP connection. Instead, proxy the child (and any
1064 subsequent ones) for TLS via a unix-domain socket channel. Logging is
1065 affected: the continued delivery log lines do not have any DNSSEC, TLS
1066 Certificate or OCSP information. TLS cipher information is still logged.
1068 JH/04 Shorten the log line for daemon startup by collapsing adjacent sets of
1069 identical IP addresses on different listening ports. Will also affect
1072 PP/02 Bug 2070: uClibc defines __GLIBC__ without providing glibc headers;
1073 add noisy ifdef guards to special-case this sillyness.
1074 Patch from Bernd Kuhls.
1076 JH/05 Tighten up the checking in isip4 (et al): dotted-quad components larger
1077 than 255 are no longer allowed.
1079 JH/06 Default openssl_options to include +no_ticket, to reduce load on peers.
1080 Disable the session-cache too, which might reduce our load. Since we
1081 currrectly use a new context for every connection, both as server and
1082 client, there is no benefit for these.
1083 GnuTLS appears to not support tickets server-side by default (we don't
1084 call gnutls_session_ticket_enable_server()) but client side is enabled
1085 by default on recent versions (3.1.3 +) unless the PFS priority string
1088 PP/03 Add $SOURCE_DATE_EPOCH support for reproducible builds, per spec at
1089 <https://reproducible-builds.org/specs/source-date-epoch/>.
1091 JH/07 Fix smtp transport use of limited max_rcpt under mua_wrapper. Previously
1092 the check for any unsuccessful recipients did not notice the limit, and
1093 erroneously found still-pending ones.
1095 JH/08 Pipeline CHUNKING command and data together, on kernels that support
1096 MSG_MORE. Only in-clear (not on TLS connections).
1098 JH/09 Avoid using a temporary file during transport using dkim. Unless a
1099 transport-filter is involved we can buffer the headers in memory for
1100 creating the signature, and read the spool data file once for the
1101 signature and again for transmission.
1103 JH/10 Enable use of sendfile in Linux builds as default. It was disabled in
1104 4.77 as the kernel support then wasn't solid, having issues in 64bit
1105 mode. Now, it's been long enough. Add support for FreeBSD also.
1107 JH/11 Bug 2104: Fix continued use of a transport connection with TLS. In the
1108 case where the routing stage had gathered several addresses to send to
1109 a host before calling the transport for the first, we previously failed
1110 to close down TLS in the old transport process before passing the TCP
1111 connection to the new process. The new one sent a STARTTLS command
1112 which naturally failed, giving a failed delivery and bloating the retry
1113 database. Investigation and fix prototype from Wolfgang Breyha.
1115 JH/12 Fix check on SMTP command input synchronisation. Previously there were
1116 false-negatives in the check that the sender had not preempted a response
1117 or prompt from Exim (running as a server), due to that code's lack of
1118 awareness of the SMTP input buffering.
1120 PP/04 Add commandline_checks_require_admin option.
1121 Exim drops privileges sanely, various checks such as -be aren't a
1122 security problem, as long as you trust local users with access to their
1123 own account. When invoked by services which pass untrusted data to
1124 Exim, this might be an issue. Set this option in main configuration
1125 AND make fixes to the calling application, such as using `--` to stop
1128 JH/13 Do pipelining under TLS. Previously, although safe, no advantage was
1129 taken. Now take care to pack both (client) MAIL,RCPT,DATA, and (server)
1130 responses to those, into a single TLS record each way (this usually means
1131 a single packet). As a side issue, smtp_enforce_sync now works on TLS
1134 PP/05 OpenSSL/1.1: use DH_bits() for more accurate DH param sizes. This
1135 affects you only if you're dancing at the edge of the param size limits.
1136 If you are, and this message makes sense to you, then: raise the
1137 configured limit or use OpenSSL 1.1. Nothing we can do for older
1140 JH/14 For the "sock" variant of the malware scanner interface, accept an empty
1141 cmdline element to get the documented default one. Previously it was
1144 JH/15 Fix a crash in the smtp transport caused when two hosts in succession
1145 are unsuable for non-message-specific reasons - eg. connection timeout,
1146 banner-time rejection.
1148 JH/16 Fix logging of delivery remote port, when specified by router, under
1151 PP/06 Repair manualroute's ability to take options in any order, even if one
1152 is the name of a transport.
1155 HS/01 Cleanup, prevent repeated use of -p/-oMr (CVE-2017-1000369)
1157 JH/17 Change the list-building routines interface to use the expanding-string
1158 triplet model, for better allocation and copying behaviour.
1160 JH/18 Prebuild the data-structure for "builtin" macros, for faster startup.
1161 Previously it was constructed the first time a possibly-matching string
1162 was met in the configuration file input during startup; now it is done
1165 JH/19 Bug 2141: Use the full-complex API for Berkeley DB rather than the legacy-
1166 compatible one, to avoid the (poorly documented) possibility of a config
1167 file in the working directory redirecting the DB files, possibly correpting
1168 some existing file. CVE-2017-10140 assigned for BDB.
1170 JH/20 Bug 2147: Do not defer for a verify-with-callout-and-random which is not
1171 cache-hot. Previously, although the result was properly cached, the
1172 initial verify call returned a defer.
1174 JH/21 Bug 2151: Avoid using SIZE on the MAIL for a callout verify, on any but
1175 the main verify for receipient in uncached-mode.
1177 JH/22 Retire historical build files to an "unsupported" subdir. These are
1178 defined as "ones for which we have no current evidence of testing".
1180 JH/23 DKIM: enforce the DNS pubkey record "h" permitted-hashes optional field,
1181 if present. Previously it was ignored.
1183 JH/24 Start using specified-initialisers in C structure init coding. This is
1184 a C99 feature (it's 2017, so now considered safe).
1186 JH/25 Use one-bit bitfields for flags in the "addr" data structure. Previously
1187 if was a fixed-sized field and bitmask ops via macros; it is now more
1190 PP/07 GitHub PR 56: Apply MariaDB build fix.
1191 Patch provided by Jaroslav Škarvada.
1193 PP/08 Bug 2161: Fix regression in sieve quoted-printable handling introduced
1194 during Coverity cleanups [4.87 JH/47]
1195 Diagnosis and fix provided by Michael Fischer v. Mollard.
1197 JH/26 Fix DKIM bug: when the pseudoheader generated for signing was exactly
1198 the right size to place the terminating semicolon on its own folded
1199 line, the header hash was calculated to an incorrect value thanks to
1200 the (relaxed) space the fold became.
1202 HS/02 Fix Bug 2130: large writes from the transport subprocess were chunked
1203 and confused the parent.
1205 JH/27 Fix SOCKS bug: an unitialized pointer was deref'd by the transport process
1206 which could crash as a result. This could lead to undeliverable messages.
1208 JH/28 Logging: "next input sent too soon" now shows where input was truncated
1211 JH/29 Fix queue_run_in_order to ignore the PID portion of the message ID. This
1212 matters on fast-turnover and PID-randomising systems, which were getting
1213 out-of-order delivery.
1215 JH/30 Fix a logging bug on aarch64: an unsafe routine was previously used for
1216 a possibly-overlapping copy. The symptom was that "Remote host closed
1217 connection in response to HELO" was logged instead of the actual 4xx
1220 JH/31 Fix CHUNKING code to properly flush the unwanted chunk after an error.
1221 Previously only that bufferd was discarded, resulting in SYMTP command
1224 JH/32 DKIM: when a message has multiple signatures matching an identity given
1225 in dkim_verify_signers, run the dkim acl once for each. Previously only
1226 one run was done. Bug 2189.
1228 JH/33 Downgrade an unfound-list name (usually a typo in the config file) from
1229 "panic the current process" to "deliberately defer". The panic log is
1230 still written with the problem list name; the mail and reject logs now
1231 get a temp-reject line for the message that was being handled, saying
1232 something like "domains check lookup or other defer". The SMTP 451
1233 message is still "Temporary local problem".
1235 JH/34 Bug 2199: Fix a use-after-free while reading smtp input for header lines.
1236 A crafted sequence of BDAT commands could result in in-use memory beeing
1237 freed. CVE-2017-16943.
1239 HS/03 Bug 2201: Fix checking for leading-dot on a line during headers reading
1240 from SMTP input. Previously it was always done; now only done for DATA
1241 and not BDAT commands. CVE-2017-16944.
1243 JH/35 Bug 2201: Flush received data in BDAT mode after detecting an error fatal
1244 to the message (such as an overlong header line). Previously this was
1245 not done and we did not exit BDAT mode. Followon from the previous item
1246 though a different problem.
1252 JH/01 Bug 1922: Support IDNA2008. This has slightly different conversion rules
1253 than -2003 did; needs libidn2 in addition to libidn.
1255 JH/02 The path option on a pipe transport is now expanded before use.
1257 PP/01 GitHub PR 50: Do not call ldap_start_tls_s on ldapi:// connections.
1258 Patch provided by "Björn", documentation fix added too.
1260 JH/03 Bug 2003: fix Proxy Protocol v2 handling: the address size field was
1261 missing a wire-to-host endian conversion.
1263 JH/04 Bug 2004: fix CHUNKING in non-PIPELINEING mode. Chunk data following
1264 close after a BDAT command line could be taken as a following command,
1265 giving a synch failure. Fix by only checking for synch immediately
1266 before acknowledging the chunk.
1268 PP/02 GitHub PR 52: many spelling fixes, which include fixing parsing of
1269 no_require_dnssec option and creation of _HAVE_TRANSPORT_APPEND_MAILDIR
1270 macro. Patches provided by Josh Soref.
1272 JH/05 Have the EHLO response advertise VRFY, if there is a vrfy ACL defined.
1273 Previously we did not; the RFC seems ambiguous and VRFY is not listed
1274 by IANA as a service extension. However, John Klensin suggests that we
1277 JH/06 Bug 2017: Fix DKIM verification in -bh test mode. The data feed into
1278 the dkim code may be unix-mode line endings rather than smtp wire-format
1279 CRLF, so prepend a CR to any bare LF.
1281 JH/07 Rationalise the coding for callout smtp conversations and transport ones.
1282 As a side-benfit, callouts can now use PIPELINING hence fewer round-trips.
1284 JH/08 Bug 2016: Fix DKIM verification vs. CHUNKING. Any BDAT commands after
1285 the first were themselves being wrongly included in the feed into dkim
1286 processing; with most chunk sizes in use this resulted in an incorrect
1287 body hash calculated value.
1289 JH/09 Bug 2014: permit inclusion of a DKIM-Signature header in a received
1290 DKIM signature block, for verification. Although advised against by
1291 standards it is specifically not ruled illegal.
1293 JH/10 Bug 2025: Fix reception of (quoted) local-parts with embedded spaces.
1295 JH/11 Bug 2029: Fix crash in DKIM verification when a message signature block is
1296 missing a body hash (the bh= tag).
1298 JH/12 Bug 2018: Re-order Proxy Protocol startup versus TLS-on-connect startup.
1299 It seems that HAProxy sends the Proxy Protocol information in clear and
1300 only then does a TLS startup, so do the same.
1302 JH/13 Bug 2027: Avoid attempting to use TCP Fast Open for non-transport client
1303 TCP connections (such as for Spamd) unless the daemon successfully set
1304 Fast Open mode on its listening sockets. This fixes breakage seen on
1305 too-old kernels or those not configured for Fast Open, at the cost of
1306 requiring both directions being enabled for TFO, and TFO never being used
1307 by non-daemon-related Exim processes.
1309 JH/14 Bug 2000: Reject messages recieved with CHUNKING but with malformed line
1310 endings, at least on the first header line. Try to canonify any that get
1311 past that check, despite the cost.
1313 JH/15 Angle-bracket nesting (an error inserted by broken sendmails) levels are
1314 now limited to an arbitrary five deep, while parsing addresses with the
1315 strip_excess_angle_brackets option enabled.
1317 PP/03 Bug 2018: For Proxy Protocol and TLS-on-connect, do not over-read and
1318 instead leave the unprompted TLS handshake in socket buffer for the
1319 TLS library to consume.
1321 PP/04 Bug 2018: Also handle Proxy Protocol v2 safely.
1323 PP/05 FreeBSD compat: handle that Ports no longer create /usr/bin/perl
1325 JH/16 Drop variables when they go out of scope. Memory management drops a whole
1326 region in one operation, for speed, and this leaves assigned pointers
1327 dangling. Add checks run only under the testsuite which checks all
1328 variables at a store-reset and panics on a dangling pointer; add code
1329 explicitly nulling out all the variables discovered. Fixes one known
1330 bug: a transport crash, where a dangling pointer for $sending_ip_address
1331 originally assigned in a verify callout, is re-used.
1333 PP/06 Drop '.' from @INC in various Perl scripts.
1335 PP/07 Switch FreeBSD iconv to always use the base-system libc functions.
1337 PP/08 Reduce a number of compilation warnings under clang; building with
1338 CC=clang CFLAGS+=-Wno-dangling-else -Wno-logical-op-parentheses
1339 should be warning-free.
1341 JH/17 Fix inbound CHUNKING when DKIM disabled at runtime.
1343 HS/01 Fix portability problems introduced by PP/08 for platforms where
1344 realloc(NULL) is not equivalent to malloc() [SunOS et al].
1346 HS/02 Bug 1974: Fix missing line terminator on the last received BDAT
1347 chunk. This allows us to accept broken chunked messages. We need a more
1348 general solution here.
1350 PP/09 Wrote util/chunking_fixqueue_finalnewlines.pl to help recover
1351 already-broken messages in the queue.
1353 JH/18 Bug 2061: Fix ${extract } corrupting an enclosing ${reduce } $value.
1355 JH/19 Fix reference counting bug in routing-generated-address tracking.
1361 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
1362 supports it and a size is available (ie. the sending peer gave us one).
1364 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
1365 years of being deprecated). The replacements are the ACLs
1366 acl_smtp_mime and acl_not_smtp_mime.
1368 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
1369 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
1370 or even an in-clear connection were permitted. Now, if the host lookup
1371 was dnssec and dane was requested then the host is only used if the
1372 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
1373 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
1374 if one fails this test.
1375 This means that a poorly-configured remote DNS will make it incommunicado;
1376 but it protects against a DNS-interception attack on it.
1378 JH/04 Bug 1810: make continued-use of an open smtp transport connection
1379 non-noisy when a race steals the message being considered.
1381 JH/05 If main configuration option tls_certificate is unset, generate a
1382 self-signed certificate for inbound TLS connections.
1384 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
1385 in rewrites and routers.
1387 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
1388 and logged a warning sing 4.83; now they are a configuration file error.
1390 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
1391 (lacking @domain). Apply the same qualification processing as RCPT.
1393 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
1395 JH/10 Support ${sha256:} applied to a string (as well as the previous
1398 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
1399 a cutthrough deliver is pending, as we always want to make a connection.
1400 This also avoids re-routing the message when later placing the cutthrough
1401 connection after a verify cache hit.
1402 Do not update it with the verify result either.
1404 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
1405 when routing results in more than one destination address.
1407 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
1408 signing (which inhibits the cutthrough capability). Previously only
1409 the presence of an option was tested; now an expansion evaluating as
1410 empty is permissible (obviously it should depend only on data available
1411 when the cutthrough connection is made).
1413 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
1414 the relevant preceding SMTP command did not note the pipelining mode.
1416 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
1417 Previously they were not counted.
1419 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
1420 as one having no matching records. Previously we deferred the message
1421 that needed the lookup.
1423 JH/17 Fakereject: previously logged as a normal message arrival "<="; now
1424 distinguished as "(=".
1426 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
1427 for missing MX records. Previously it only worked for missing A records.
1429 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
1431 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
1432 after the data-go-ahead and data-ack. Patch from Jason Betts.
1434 JH/21 Bug 1846: Send DMARC forensic reports for reject and quarantine results,
1435 even for a "none" policy. Patch from Tony Meyer.
1437 JH/22 Fix continued use of a connection for further deliveries. If a port was
1438 specified by a router, it must also match for the delivery to be
1441 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
1442 When one of the recipients of a message was unsuitable for the connection
1443 (has no matching addresses), we lost track of needing to mark it
1444 deferred. As a result mail would be lost.
1446 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
1448 JH/25 Decoding ACL controls is now done using a binary search; the source code
1449 takes up less space and should be simpler to maintain. Merge the ACL
1450 condition decode tables also, with similar effect.
1452 JH/26 Fix problem with one_time used on a redirect router which returned the
1453 parent address unchanged. A retry would see the parent address marked as
1454 delivered, so not attempt the (identical) child. As a result mail would
1457 JH/27 Fix a possible security hole, wherein a process operating with the Exim
1458 UID can gain a root shell. Credit to http://www.halfdog.net/ for
1459 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
1462 JH/28 Enable {spool,log} filesystem space and inode checks as default.
1463 Main config options check_{log,spool}_{inodes,space} are now
1464 100 inodes, 10MB unless set otherwise in the configuration.
1466 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
1467 Previously it only applied to the main-section connection policy
1470 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
1472 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
1473 by me. Added RFC7919 DH primes as an alternative.
1475 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
1476 are not in the system include path.
1478 JH/31 Fix longstanding bug with aborted TLS server connection handling. Under
1479 GnuTLS, when a session startup failed (eg because the client disconnected)
1480 Exim did stdio operations after fclose. This was exposed by a recent
1481 change which nulled out the file handle after the fclose.
1483 JH/32 Bug 1909: Fix OCSP proof verification for cases where the proof is
1484 signed directly by the cert-signing cert, rather than an intermediate
1485 OCSP-signing cert. This is the model used by LetsEncrypt.
1487 JH/33 Bug 1914: Ensure socket is nonblocking before draining after SMTP QUIT.
1489 HS/01 Fix leak in verify callout under GnuTLS, about 3MB per recipient on
1490 an incoming connection.
1492 HS/02 Bug 1802: Do not half-close the connection after sending a request
1495 HS/03 Use "auto" as the default EC curve parameter. For OpenSSL < 1.0.2
1496 fallback to "prime256v1".
1498 JH/34 SECURITY: Use proper copy of DATA command in error message.
1499 Could leak key material. Remotely exploitable. CVE-2016-9963.
1505 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
1506 and 3.4.4 - once the server is enabled to respond to an OCSP request
1507 it does even when not requested, resulting in a stapling non-aware
1508 client dropping the TLS connection.
1510 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
1511 support variable-length bit vectors. No functional change.
1513 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
1514 The I= interface field on outgoing lines is now after the H= remote
1515 host field, same as incoming lines. There is a separate
1516 outgoing_interface log selector which allows you to disable the
1519 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
1520 If not running log_selector +smtp_connection the mainlog would be held
1521 open indefinitely after a "too many connections" event, including to a
1522 deleted file after a log rotate. Leave the per net connection logging
1523 leaving it open for efficiency as that will be quickly detected by the
1524 check on the next write.
1526 HS/01 Bug 1671: Fix post transport crash.
1527 Processing the wait-<transport> messages could crash the delivery
1528 process if the message IDs didn't exist for some reason. When
1529 using 'split_spool_directory=yes' the construction of the spool
1530 file name failed already, exposing the same netto behaviour.
1532 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
1533 mime_regex ACL conditions.
1535 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
1536 to DSN fail messages (bounces): remote IP, remote greeting, remote response
1537 to HELO, local diagnostic string.
1539 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
1540 log line to debug. Even when configured with a tls authenticator many
1541 client connections are expected to not authenticate in this way, so
1542 an authenticate fail is not an error.
1544 HS/02 Add the Exim version string to the process info. This way exiwhat
1545 gives some more detail about the running daemon.
1547 JH/06 Bug 1395: time-limit caching of DNS lookups, to the TTL value. This may
1548 matter for fast-change records such as DNSBLs.
1550 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
1551 retry record, even if constant. There may be multiple transports with
1552 different interface settings and the retry behaviour needs to be kept
1555 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
1557 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
1559 JH/10 Bug 840: fix log_defer_output option of pipe transport
1561 JH/11 Bug 830: use same host for all RCPTS of a message, even under
1562 hosts_randomize. This matters a lot when combined with mua_wrapper.
1564 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
1565 ${quote_pgsql:<string>} operator.
1567 JH/13 Bug 1708: avoid misaligned access in cached lookup.
1569 JH/14 Change header file name for freeradius-client. Relevant if compiling
1570 with Radius support; from the Gentoo tree and checked under Fedora.
1572 JH/15 Bug 1712: Introduce $prdr_requested flag variable
1574 JH/16 Bug 1714: Permit an empty string as expansion result for transport
1575 option transport_filter, meaning no filtering.
1577 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
1579 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
1580 defaults to "*" (all hosts). The variable is now available when not built
1581 with TLS, default unset, mainly to enable keeping the testsuite sane.
1582 If a server certificate is not supplied (via tls_certificate) an error is
1583 logged, and clients will find TLS connections fail on startup. Presumably
1584 they will retry in-clear.
1585 Packagers of Exim are strongly encouraged to create a server certificate
1586 at installation time.
1588 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
1589 with the $config_file variable.
1591 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
1592 in transport context, after the attempt, and per-recipient. The latter type
1593 is per host attempted. The event data is the error message, and the errno
1594 information encodes the lookup type (A vs. MX) used for the (first) host,
1595 and the trailing two digits of the smtp 4xx response.
1597 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
1598 to write to mainlog (or rejectlog, paniclog) in the window between file
1599 creation and permissions/ownership being changed. Particularly affects
1600 installations where exicyclog is run as root, rather than exim user;
1601 result is that the running daemon panics and dies.
1603 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
1605 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
1606 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
1607 "pri" and "weight". Note that the previous implicit priority given by the
1608 list order is no longer honoured.
1610 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalization
1611 for DKIM processing.
1613 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
1614 by defining SUPPORT_SOCKS.
1616 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
1617 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
1618 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
1619 variables are renamed to proxy_{local,external}_{address,port}.
1621 JH/27 Move Internationalisation support from Experimental to mainline, enabled
1622 for a build by defining SUPPORT_I18N
1624 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
1625 of the query string, and make ${quote_redis:} do that quoting.
1627 JH/29 Move Events support from Experimental to mainline, enabled by default
1628 and removable for a build by defining DISABLE_EVENT.
1630 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
1632 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
1633 cached by the daemon.
1635 JH/32 Move Redis support from Experimental to mainline, enabled for a build
1636 by defining LOOKUP_REDIS. The libhiredis library is required.
1638 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
1639 keys are given for lookup.
1641 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
1642 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
1643 only supported when built with TLS support. The PolarSSL SHA routines
1644 are still used when the TLS library is too old for convenient support.
1646 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
1647 openssl_options), for security. OpenSSL forces this from version 1.1.0
1648 server-side so match that on older versions.
1650 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
1651 allocation for $value could be released as the expansion processing
1652 concluded, but leaving the global pointer active for it.
1654 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
1655 and to use the domains and local_parts ACL conditions.
1657 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
1658 incorrectly not doubled on cutthrough transmission, hence seen as a
1659 body-termination at the receiving system - resulting in truncated mails.
1660 Commonly the sender saw a TCP-level error, and retransmitted the message
1661 via the normal store-and-forward channel. This could result in duplicates
1662 received - but deduplicating mailstores were liable to retain only the
1663 initial truncated version.
1665 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
1667 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
1669 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
1670 we're in there, support oversigning also; bug 1309.
1672 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
1674 HS/04 Add support for keep_environment and add_environment options.
1676 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
1677 either intentional arithmetic overflow during PRNG, or testing config-
1680 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
1681 delivery resulted in actual delivery. Cancel cutthrough before DATA
1684 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
1685 rejects a recipient: pass the reject to the originator.
1687 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
1688 Many were false-positives and ignorable, but it's worth fixing the
1691 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
1692 for the new environment-manipulation done at startup. Move the routines
1693 from being local to tls.c to being global via the os.c file.
1695 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
1696 an extract embedded as result-arg for a map, the first arg for extract
1697 is unavailable so we cannot tell if this is a numbered or keyed
1698 extraction. Accept either.
1704 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
1707 JH/02 The smtp transport option "multi_domain" is now expanded.
1709 JH/03 The smtp transport now requests PRDR by default, if the server offers
1712 JH/04 Certificate name checking on server certificates, when exim is a client,
1713 is now done by default. The transport option tls_verify_cert_hostnames
1714 can be used to disable this per-host. The build option
1715 EXPERIMENTAL_CERTNAMES is withdrawn.
1717 JH/05 The value of the tls_verify_certificates smtp transport and main options
1718 default to the word "system" to access the system default CA bundle.
1719 For GnuTLS, only version 3.0.20 or later.
1721 JH/06 Verification of the server certificate for a TLS connection is now tried
1722 (but not required) by default. The verification status is now logged by
1723 default, for both outbound TLS and client-certificate supplying inbound
1726 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
1729 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
1730 Status Notification (bounce) messages are now MIME format per RFC 3464.
1731 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
1732 under the control of the dsn_advertise_hosts option, and routers may
1733 have a dsn_lasthop option.
1735 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
1736 default, modifiable by a malware= option. The list separator for
1737 the options can now be changed in the usual way. Bug 68.
1739 JH/10 The smtp_receive_timeout main option is now expanded before use.
1741 JH/11 The incoming_interface log option now also enables logging of the
1742 local interface on delivery outgoing connections.
1744 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
1745 if the interface and destination host and port all match.
1747 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
1750 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
1751 Patch from Andrew Lewis.
1753 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
1754 now supports optional time-restrictions, weighting, and priority
1755 modifiers per server. Patch originally by <rommer@active.by>.
1757 JH/16 The spamd_address main option now supports a mixed list of local
1758 and remote servers. Remote servers can be IPv6 addresses, and
1759 specify a port-range.
1761 JH/17 Bug 68: The spamd_address main option now supports an optional
1762 timeout value per server.
1764 JH/18 Bug 1581: Router and transport options headers_add/remove can
1765 now have the list separator specified.
1767 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
1770 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
1773 JH/21 Support for the A6 type of dns record is withdrawn.
1775 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
1776 rather than the verbs used.
1778 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
1779 from 255 to 1024 chars.
1781 JH/24 Verification callouts now attempt to use TLS by default.
1783 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
1784 are generic router options now. The defaults didn't change.
1786 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
1787 Original patch from Alexander Shikoff, worked over by JH.
1789 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
1790 files if ZCAT_COMMAND is not executable.
1792 JH/26 Bug 1539: Add timeout/retry options on dnsdb lookups.
1794 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
1796 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
1797 Normally benign, it bites when the pair was led to by a CNAME;
1798 modern usage is to not canonicalize the domain to a CNAME target
1799 (and we were inconsistent anyway for A-only vs AAAA+A).
1801 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
1803 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
1804 when evaluating $sender_host_dnssec.
1806 JH/31 Check the HELO verification lookup for DNSSEC, adding new
1807 $sender_helo_dnssec variable.
1809 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
1811 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
1813 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
1815 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
1816 documented as working, but never had. Support all but $spam_report.
1818 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
1819 added for tls authenticator.
1821 HS/03 Add perl_taintmode main config option
1827 TL/01 When running the test suite, the README says that variables such as
1828 no_msglog_check are global and can be placed anywhere in a specific
1829 test's script, however it was observed that placement needed to be near
1830 the beginning for it to behave that way. Changed the runtest perl
1831 script to read through the entire script once to detect and set these
1832 variables, reset to the beginning of the script, and then run through
1833 the script parsing/test process like normal.
1835 TL/02 The BSD's have an arc4random API. One of the functions to induce
1836 adding randomness was arc4random_stir(), but it has been removed in
1837 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
1838 function when detected.
1840 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
1841 cause callback expansion.
1843 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
1844 syntax errors in an expansion can be treated as a string instead of
1845 logging or causing an error, due to the internal use of bool_lax
1846 instead of bool when processing it.
1848 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
1849 server certificates when making smtp deliveries.
1851 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
1853 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
1855 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
1857 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
1858 Merged patch from Sebastian Wiedenroth.
1860 JH/05 Fix results-pipe from transport process. Several recipients, combined
1861 with certificate use, exposed issues where response data items split
1862 over buffer boundaries were not parsed properly. This eventually
1863 resulted in duplicates being sent. This issue only became common enough
1864 to notice due to the introduction of connection certificate information,
1865 the item size being so much larger. Found and fixed by Wolfgang Breyha.
1867 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
1868 size buffer was used, resulting in syntax errors when an expansion
1871 JH/07 Add support for directories of certificates when compiled with a GnuTLS
1872 version 3.3.6 or later.
1874 JH/08 Rename the TPDA experimental facility to Event Actions. The #ifdef
1875 is EXPERIMENTAL_EVENT, the main-configuration and transport options
1876 both become "event_action", the variables become $event_name, $event_data
1877 and $event_defer_errno. There is a new variable $verify_mode, usable in
1878 routers, transports and related events. The tls:cert event is now also
1879 raised for inbound connections, if the main configuration event_action
1882 TL/06 In test suite, disable OCSP for old versions of openssl which contained
1883 early OCSP support, but no stapling (appears to be less than 1.0.0).
1885 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
1886 server certificate names available under the smtp transport option
1887 "tls_verify_cert_hostname" now do not permit multi-component wildcard
1890 JH/10 Time-related extraction expansions from certificates now use the main
1891 option "timezone" setting for output formatting, and are consistent
1892 between OpenSSL and GnuTLS compilations. Bug 1541.
1894 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
1895 encoded parameter in the incoming message. Bug 1558.
1897 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
1898 include certificate info, eximon was claiming there were spoolfile
1901 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
1903 JH/14 Log delivery-related information more consistently, using the sequence
1904 "H=<name> [<ip>]" wherever possible.
1906 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
1907 are problematic for Debian distribution, omit them from the release
1910 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
1912 JH/16 Fix string representation of time values on 64bit time_t architectures.
1915 JH/17 Fix a null-indirection in certextract expansions when a nondefault
1916 output list separator was used.
1921 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
1922 checkers that were complaining about end of non-void function with no
1925 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
1926 This was a regression introduced in 4.83 by another bugfix.
1928 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
1930 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
1931 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
1937 TF/01 Correctly close the server side of TLS when forking for delivery.
1939 When a message was received over SMTP with TLS, Exim failed to clear up
1940 the incoming connection properly after forking off the child process to
1941 deliver the message. In some situations the subsequent outgoing
1942 delivery connection happened to have the same fd number as the incoming
1943 connection previously had. Exim would try to use TLS and fail, logging
1944 a "Bad file descriptor" error.
1946 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
1947 utilities have not been installed.
1949 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
1950 temporary space as the ACL may create new global variables.
1952 TL/01 LDAP support uses per connection or global context settings, depending
1953 upon the detected version of the libraries at build time.
1955 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
1956 to extract and use the src ip:port in logging and expansions as if it
1957 were a direct connection from the outside internet. PPv2 support was
1958 updated based on HAProxy spec change in May 2014.
1960 JH/02 Add ${listextract {number}{list}{success}{fail}}.
1962 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
1963 Properly escape header and check for NULL return.
1965 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
1968 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
1970 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
1971 characters in header names, implemented as a verify condition.
1972 Contributed by Michael Fischer v. Mollard.
1974 TL/05 Rename SPF condition results err_perm and err_temp to standardized
1975 results permerror and temperror. Previous values are deprecated but
1976 still accepted. In a future release, err_perm and err_temp will be
1977 completely removed, which will be a backward incompatibility if the
1978 ACL tests for either of these two old results. Patch contributed by
1979 user bes-internal on the mailing list.
1981 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
1983 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
1984 selectors, in both main and reject logs.
1986 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
1989 JH/07 Add malware type "sock" for talking to simple daemon.
1991 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
1993 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
1994 routers/transports under cutthrough routing.
1996 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
1997 numbers. Touch up "bool" conditional to keep the same definition.
1999 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
2001 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
2003 JH/12 Expand items in router/transport headers_add or headers_remove lists
2004 individually rather than the list as a whole. Bug 1452.
2006 Required for reasonable handling of multiple headers_ options when
2007 they may be empty; requires that headers_remove items with embedded
2008 colons must have them doubled (or the list-separator changed).
2010 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
2011 view the policy declared in the DMARC record. Currently, $dmarc_status
2012 is a combined value of both the record presence and the result of the
2015 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
2017 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
2018 dnslookup router and the smtp transport (applying to the forward
2021 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
2022 of ldap servers used for a specific lookup. Patch provided by Heiko
2025 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
2026 New variable $lookup_dnssec_authenticated for observability.
2028 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
2029 Patch submitted by Lars Timman.
2031 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
2033 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
2034 Requires trusted mode and valid format message id, aborts otherwise.
2035 Patch contributed by Heiko Schlichting.
2037 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
2038 certextract with support for various fields. Bug 1358.
2040 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
2041 is requested by default, modifiable by smtp transport option
2044 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
2045 operate on certificate variables to give certificate fingerprints
2046 Also new ${sha256:cert_variable}.
2048 JH/23 The PRDR feature is moved from being Experimental into the mainline.
2050 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
2051 Christian Aistleitner.
2053 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
2055 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
2056 file. Patch from Wolfgang Breyha.
2058 JH/25 Expand the coverage of the delivery $host and $host_address to
2059 client authenticators run in verify callout. Bug 1476.
2061 JH/26 Port service names are now accepted for tls_on_connect_ports, to
2062 align with daemon_smtp_ports. Bug 72.
2064 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
2065 support and error reporting did not work properly.
2067 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
2068 and is readable. Patch from Andrew Colin Kissa.
2070 TL/14 Enhance documentation of ${run expansion and how it parses the
2071 commandline after expansion, particularly in the case when an
2072 unquoted variable expansion results in an empty value.
2074 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
2076 PP/02 Fix internal collision of T_APL on systems which support RFC3123
2077 by renaming away from it. Addresses GH issue 15, reported by
2080 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
2082 TL/15 SECURITY: prevent double expansion in math comparison functions
2083 (can expand unsanitized data). Not remotely exploitable.
2090 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
2092 PP/02 Make -n do something, by making it not do something.
2093 When combined with -bP, the name of an option is not output.
2095 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
2098 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
2099 $sender_host_name and config options to manage this, and basic check
2102 PP/05 DSCP support for outbound connections and control modifier for inbound.
2104 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
2105 (Only plugin which currently uses this is kerberos4, which nobody should
2106 be using, but we should make it available and other future plugins might
2107 conceivably use it, even though it would break NAT; stuff *should* be
2108 using channel bindings instead).
2110 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
2111 name; added for Sendmail compatibility; requires admin caller.
2112 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
2113 just ignore it); requires trusted caller.
2114 Also parse but ignore: -Ac -Am -X<logfile>
2117 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
2119 TL/02 Add +smtp_confirmation as a default logging option.
2121 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
2122 Patch by Magnus Holmgren from 2007-02-20.
2124 TL/04 Bugzilla 1281 - Spec typo.
2125 Bugzilla 1283 - Spec typo.
2126 Bugzilla 1290 - Spec grammar fixes.
2128 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
2130 TL/06 Add Experimental DMARC support using libopendmarc libraries.
2132 TL/07 Fix an out of order global option causing a segfault. Reported to dev
2133 mailing list by by Dmitry Isaikin.
2135 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
2137 JH/02 Support "G" suffix to numbers in ${if comparisons.
2139 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
2141 NM/01 Bugzilla 1197 - Spec typo
2142 Bugzilla 1196 - Spec examples corrections
2144 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
2146 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
2147 gnutls_enable_pkcs11, but renamed to more accurately indicate its
2150 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
2151 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
2153 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
2154 "acl {{name}{arg}...}", and optional args on acl condition
2157 JH/05 Permit multiple router/transport headers_add/remove lines.
2159 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
2161 JH/07 Avoid using a waiting database for a single-message-only transport.
2162 Performance patch from Paul Fisher. Bugzilla 1262.
2164 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
2167 JH/09 Add $headers_added variable, with content from use of ACL modifier
2168 add_header (but not yet added to the message). Bugzilla 199.
2170 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
2171 Pulled from Bugzilla 817 by Wolfgang Breyha.
2173 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
2175 (nb: this is the same fix as in Exim 4.80.1)
2177 JH/11 Add A= logging on delivery lines, and a client_set_id option on
2180 JH/12 Add optional authenticated_sender logging to A= and a log_selector
2183 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
2185 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
2186 advertise SMTP AUTH mechanism to us, instead of a generic
2187 protocol violation error. Also, make Exim more robust to bad
2188 data from the Dovecot auth socket.
2190 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
2192 When a queue runner is handling a message, Exim first routes the
2193 recipient addresses, during which it prunes them based on the retry
2194 hints database. After that it attempts to deliver the message to
2195 any remaining recipients. It then updates the hints database using
2198 So if a recipient address works intermittently, it can get repeatedly
2199 deferred at routing time. The retry hints record remains fresh so the
2200 address never reaches the final cutoff time.
2202 This is a fairly common occurrence when a user is bumping up against
2203 their storage quota. Exim had some logic in its local delivery code
2204 to deal with this. However it did not apply to per-recipient defers
2205 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
2207 This change adds a proper retry rule check during routing so that the
2208 final cutoff time is checked against the message's age. We only do
2209 this check if there is an address retry record and there is not a
2210 domain retry record; this implies that previous attempts to handle
2211 the address had the retry_use_local_parts option turned on. We use
2212 this as an approximation for the destination being like a local
2213 delivery, as in LMTP.
2215 I suspect this new check makes the old local delivery cutoff check
2216 redundant, but I have not verified this so I left the code in place.
2218 TF/02 Correct gecos expansion when From: is a prefix of the username.
2220 Test 0254 submits a message to Exim with the header
2224 When I ran the test suite under the user fanf2, Exim expanded
2225 the header to contain my full name, whereas it should have added
2226 a Resent-Sender: header. It erroneously treats any prefix of the
2227 username as equal to the username.
2229 This change corrects that bug.
2231 GF/01 DCC debug and logging tidyup
2232 Error conditions log to paniclog rather than rejectlog.
2233 Debug lines prefixed by "DCC: " to remove any ambiguity.
2235 TF/03 Avoid unnecessary rebuilds of lookup-related code.
2237 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
2238 Bug spotted by Jeremy Harris; was flawed since initial commit.
2239 Would have resulted in OCSP responses post-SNI triggering an Exim
2240 NULL dereference and crash.
2242 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
2244 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
2245 Bug detection, analysis and fix by Samuel Thibault.
2246 Bugzilla 1331, Debian bug #698092.
2248 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
2250 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
2251 Server implementation by Todd Lyons, client by JH.
2252 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
2253 config variable "prdr_enable" controls whether the server
2254 advertises the facility. If the client requests PRDR a new
2255 acl_data_smtp_prdr ACL is called once for each recipient, after
2256 the body content is received and before the acl_smtp_data ACL.
2257 The client is controlled by both of: a hosts_try_prdr option
2258 on the smtp transport, and the server advertisement.
2259 Default client logging of deliveries and rejections involving
2260 PRDR are flagged with the string "PRDR".
2262 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
2263 fclose(). Diagnosis by Todd Lyons.
2265 PP/17 Update configure.default to handle IPv6 localhost better.
2266 Patch by Alain Williams (plus minor tweaks).
2269 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
2270 This is now consistent with GnuTLS, and is now documented: the
2271 previous undocumented portable approach to treating the option as
2272 unset was to force an expansion failure. That still works, and
2273 an empty string is now equivalent.
2275 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
2276 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
2277 not performing validation itself.
2279 PP/20 Added force_command boolean option to pipe transport.
2280 Patch from Nick Koston, of cPanel Inc.
2282 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
2285 TF/04 Added udpsend ACL modifier and hexquote expansion operator
2287 PP/21 Fix eximon continuous updating with timestamped log-files.
2288 Broken in a format-string cleanup in 4.80, missed when I repaired the
2289 other false fix of the same issue.
2290 Report and fix from Heiko Schlichting.
2293 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
2294 Report from Prashanth Katuri.
2296 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
2297 It's SecureTransport, so affects any MacOS clients which use the
2298 system-integrated TLS libraries, including email clients.
2300 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
2301 using a MIME ACL for non-SMTP local injection.
2302 Report and assistance in diagnosis by Warren Baker.
2304 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
2306 JH/16 Fix comparisons for 64b. Bugzilla 1385.
2308 TL/09 Add expansion variable $authenticated_fail_id to keep track of
2309 last id that failed so it may be referenced in subsequent ACL's.
2311 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
2314 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
2315 ldap library initialization, allowing self-signed CA's to be
2316 used. Also properly sets require_cert option later in code by
2317 using NULL (global ldap config) instead of ldap handle (per
2318 session). Bug diagnosis and testing by alxgomz.
2320 TL/12 Enhanced documentation in the ratelimit.pl script provided in
2321 the src/util/ subdirectory.
2323 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
2324 renamed to Transport Post Delivery Action by Jeremy Harris, as
2327 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
2328 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
2329 redis_servers = needs to be configured which will be used by the redis
2330 lookup. Patch from Warren Baker, of The Packet Hub.
2332 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
2334 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
2335 hostname or reverse DNS when processing a host list. Used suggestions
2336 from multiple comments on this bug.
2338 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
2340 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
2341 Missed a few lines, added it to make the runtest require no keyboard
2344 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
2345 contains upper case chars. Make router use caseful_local_part.
2347 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
2348 support when GnuTLS has been built with p11-kit.
2354 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
2356 This, or similar/improved, will also be change PP/11 of 4.82.
2362 PP/01 Handle short writes when writing local log-files.
2363 In practice, only affects FreeBSD (8 onwards).
2364 Bugzilla 1053, with thanks to Dmitry Isaikin.
2366 NM/01 Bugzilla 949 - Documentation tweak
2368 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
2371 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
2373 PP/02 Implemented gsasl authenticator.
2375 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
2377 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
2378 `pkg-config foo` for cflags/libs.
2380 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
2381 with rest of GSASL and with heimdal_gssapi.
2383 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
2384 `pkg-config foo` for cflags/libs for the TLS implementation.
2386 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
2387 properties get this fed in as external SSF. A number of robustness
2388 and debugging improvements to the cyrus_sasl authenticator.
2390 PP/08 cyrus_sasl server now expands the server_realm option.
2392 PP/09 Bugzilla 1214 - Log authentication information in reject log.
2393 Patch by Jeremy Harris.
2395 PP/10 Added dbmjz lookup type.
2397 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
2399 PP/12 MAIL args handles TAB as well as SP, for better interop with
2400 non-compliant senders.
2401 Analysis and variant patch by Todd Lyons.
2403 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
2404 Bug report from Lars Müller <lars@samba.org> (via SUSE),
2405 Patch from Dirk Mueller <dmueller@suse.com>
2407 PP/13 tls_peerdn now print-escaped for spool files.
2408 Observed some $tls_peerdn in wild which contained \n, which resulted
2409 in spool file corruption.
2411 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
2412 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
2413 or write after TLS renegotiation, which otherwise led to messages
2416 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
2417 as a tracking header (ie: a signed header comes before the signature).
2418 Patch from Wolfgang Breyha.
2420 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
2421 comma-sep list; embedded commas doubled.
2423 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
2425 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
2427 Report and patch from Dmitry Banschikov.
2429 PP/16 Removed "dont_insert_empty_fragments" from "openssl_options".
2430 Removed SSL_clear() after SSL_new() which led to protocol negotiation
2431 failures. We appear to now support TLS1.1+ with Exim.
2433 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
2434 lets Exim select keys and certificates based upon TLS SNI from client.
2435 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
2436 before an outbound SMTP session. New log_selector, +tls_sni.
2438 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
2439 NULL dereference. Report and patch from Alun Jones.
2441 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
2442 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
2443 Not seeing resolver debug output on NetBSD, but suspect this is a
2444 resolver implementation change.
2446 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
2447 Left warnings. Added "eximon gdb" invocation mode.
2449 PP/21 Defaulting "accept_8bitmime" to true, not false.
2451 PP/22 Added -bw for inetd wait mode support.
2453 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
2454 locate the relevant includes and libraries. Made this the default.
2456 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
2457 Bugzilla 1246, report and most of solution from Tomasz Kusy.
2459 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
2460 This may cause build issues on older platforms.
2462 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
2463 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
2464 gnutls_require_mac & gnutls_require_protocols (no longer supported).
2465 Added SNI support via GnuTLS too.
2466 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
2468 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
2470 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
2471 Applied second patch from Janne, implementing suggestion to default
2472 multiple-strings-in-record handling to match SPF spec.
2474 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
2476 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
2477 read-only, out of scope).
2478 Patch from Wolfgang Breyha, report from Stuart Northfield.
2480 PP/29 Fix three issues highlighted by clang analyser static analysis.
2481 Only crash-plausible issue would require the Cambridge-specific
2482 iplookup router and a misconfiguration.
2483 Report from Marcin Mirosław.
2485 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
2487 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
2488 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
2489 As part of this, removing so much warning spew let me fix some minor
2490 real issues in debug logging.
2492 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
2493 assignment on my part. Fixed.
2495 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
2496 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
2497 Janne Snabb (who went above and beyond: thank you).
2499 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
2500 string otherwise requires a connection and a bunch more work and it's
2501 relatively easy to get wrong. Should also expose TLS library linkage
2504 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
2505 64-bit ${eval} (JH/03).
2507 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
2508 GNU libc to support some of the 64-bit stuff, should not lead to
2509 conflicts. Defined before os.h is pulled in, so if a given platform
2510 needs to override this, it can.
2512 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
2513 protection layer was required, which is not implemented.
2514 Bugzilla 1254, patch from Wolfgang Breyha.
2516 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
2517 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
2518 tls_dhparam take prime identifiers. Also unbreak combination of
2519 OpenSSL+DH_params+TLSSNI.
2521 PP/39 Disable SSLv2 by default in OpenSSL support.
2527 PP/01 Solaris build fix for Oracle's LDAP libraries.
2528 Bugzilla 1109, patch from Stephen Usher.
2530 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
2532 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
2535 TF/02 Fix a couple more cases where we did not log the error message
2536 when unlink() failed. See also change 4.74-TF/03.
2538 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
2539 lock up or crash if it happened to be inside a call to libc when it
2540 got a SIGUSR1 from exiwhat.
2542 The SIGUSR1 handler appends the current process status to the process
2543 log which is later printed by exiwhat. It used to use the general
2544 purpose logging code to do this, but several functions it calls are
2545 not safe for signals.
2547 The new output code in the SIGUSR1 handler is specific to the process
2548 log, and simple enough that it's easy to inspect for signal safety.
2549 Removing some special cases also simplifies the general logging code.
2550 Removing the spurious timestamps from the process log simplifies
2553 TF/04 Improved ratelimit ACL condition.
2555 The /noupdate option has been deprecated in favour of /readonly which
2556 has clearer semantics. The /leaky, /strict, and /readonly update modes
2557 are mutually exclusive. The update mode is no longer included in the
2558 database key; it just determines when the database is updated. (This
2559 means that when you upgrade Exim will forget old rate measurements.)
2561 Exim now checks that the per_* options are used with an update mode that
2562 makes sense for the current ACL. For example, when Exim is processing a
2563 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
2564 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
2565 must specify per_mail/readonly. If you omit the update mode it defaults to
2566 /leaky where that makes sense (as before) or /readonly where required.
2568 The /noupdate option is now undocumented but still supported for
2569 backwards compatibility. It is equivalent to /readonly except that in
2570 ACLs where /readonly is required you may specify /leaky/noupdate or
2571 /strict/noupdate which are treated the same as /readonly.
2573 A useful new feature is the /count= option. This is a generalization
2574 of the per_byte option, so that you can measure the throughput of other
2575 aggregate values. For example, the per_byte option is now equivalent
2576 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
2578 The per_rcpt option has been generalized using the /count= mechanism
2579 (though it's more complicated than the per_byte equivalence). When it is
2580 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
2581 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
2582 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
2583 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
2584 Note that using per_rcpt with a non-readonly update mode in more than
2585 one ACL will cause the recipients to be double-counted. (The per_mail
2586 and per_byte options don't have this problem.)
2588 The handling of very low rates has changed slightly. If the computed rate
2589 is less than the event's count (usually one) then this event is the first
2590 after a long gap. In this case the rate is set to the same as this event's
2591 count, so that the first message of a spam run is counted properly.
2593 The major new feature is a mechanism for counting the rate of unique
2594 events. The new per_addr option counts the number of different
2595 recipients that someone has sent messages to in the last time period. It
2596 behaves like per_rcpt if all the recipient addresses are different, but
2597 duplicate recipient addresses do not increase the measured rate. Like
2598 the /count= option this is a general mechanism, so the per_addr option
2599 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
2600 example, measure the rate that a client uses different sender addresses
2601 with the options per_mail/unique=$sender_address. There are further
2602 details in the main documentation.
2604 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
2606 TF/06 Removed a few PCRE remnants.
2608 TF/07 Automatically extract Exim's version number from tags in the git
2609 repository when doing development or release builds.
2611 PP/02 Raise smtp_cmd_buffer_size to 16kB.
2612 Bugzilla 879. Patch from Paul Fisher.
2614 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
2615 Heavily based on revision 40f9a89a from Simon Arlott's tree.
2618 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
2620 PP/05 Variable $av_failed, true if the AV scanner deferred.
2621 Bugzilla 1078. Patch from John Horne.
2623 PP/06 Stop make process more reliably on build failure.
2624 Bugzilla 1087. Patch from Heiko Schlittermann.
2626 PP/07 Make maildir_use_size_file an _expandable_ boolean.
2627 Bugzilla 1089. Patch from Heiko Schlittermann.
2629 PP/08 Handle ${run} returning more data than OS pipe buffer size.
2630 Bugzilla 1131. Patch from Holger Weiß.
2632 PP/09 Handle IPv6 addresses with SPF.
2633 Bugzilla 860. Patch from Wolfgang Breyha.
2635 PP/10 GnuTLS: support TLS 1.2 & 1.1.
2637 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
2640 PP/11 match_* no longer expand right-hand-side by default.
2641 New compile-time build option, EXPAND_LISTMATCH_RHS.
2642 New expansion conditions, "inlist", "inlisti".
2644 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
2646 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
2648 PP/14 fix log_write() format string regression from TF/03.
2649 Bugzilla 1152. Patch from Dmitry Isaikin.
2655 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
2657 PP/02 Harmonised TLS library version reporting; only show if debugging.
2658 Layout now matches that introduced for other libraries in 4.74 PP/03.
2660 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
2662 PP/04 New "dns_use_edns0" global option.
2664 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
2667 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
2668 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
2670 TK/01 Updated PolarSSL code to 0.14.2.
2671 Bugzilla 1097. Patch from Andreas Metzler.
2673 PP/07 Catch divide-by-zero in ${eval:...}.
2674 Fixes bugzilla 1102.
2676 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
2679 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
2680 format-string attack -- SECURITY: remote arbitrary code execution.
2682 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
2683 time unintentionally subject to list matching rules, letting the header
2684 cause arbitrary Exim lookups (of items which can occur in lists, *not*
2685 arbitrary string expansion). This allowed for information disclosure.
2687 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
2688 INT_MIN/-1 -- value coerced to INT_MAX.
2694 NM/01 Workaround for PCRE version dependency in version reporting
2697 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
2698 This fixes portability to compilers other than gcc, notably
2699 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
2701 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
2702 makefiles for portability to HP-UX and POSIX correctness.
2704 PP/01 Permit LOOKUP_foo enabling on the make command-line.
2705 Also via indented variable definition in the Makefile.
2706 (Debugging by Oliver Heesakkers).
2708 PP/02 Restore caching of spamd results with expanded spamd_address.
2709 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
2711 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
2712 Improves build reliability. Fix from: Frank Elsner
2714 NM/02 Fix wide character breakage in the rfc2047 coding
2715 Fixes bug 1064. Patch from Andrey N. Oktyabrski
2717 NM/03 Allow underscore in dnslist lookups
2718 Fixes bug 1026. Patch from Graeme Fowler
2720 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
2721 Code patches from Adam Ciarcinski of NetBSD.
2723 NM/04 Fixed exiqgrep to cope with mailq missing size issue
2726 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
2727 is logged, to avoid truncation. Patch from John Horne.
2729 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
2730 Patch from Jakob Hirsch.
2732 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
2733 SQL string expansion failure details.
2734 Patch from Andrey Oktyabrski.
2736 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
2737 Patch from Simon Arlott.
2739 PP/09 New lookups functionality failed to compile on old gcc which rejects
2740 extern declarations in function scope.
2741 Patch from Oliver Fleischmann
2743 PP/10 Use sig_atomic_t for flags set from signal handlers.
2744 Check getgroups() return and improve debugging.
2745 Fixed developed for diagnosis in bug 927 (which turned out to be
2748 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
2749 Patch from Mark Zealey.
2751 PP/12 Bugzilla 1056: Improved spamd server selection.
2752 Patch from Mark Zealey.
2754 PP/13 Bugzilla 1086: Deal with maildir quota file races.
2755 Based on patch from Heiko Schlittermann.
2757 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
2758 Patch from Uwe Doering, sign-off by Michael Haardt.
2760 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
2761 variable declaration deep within a block. Bug and patch from
2764 PP/15 lookups-Makefile IRIX compatibility coercion.
2766 PP/16 Make DISABLE_DKIM build knob functional.
2768 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
2769 Patch by Simon Arlott
2771 TF/03 Fix valgrind.h portability to C89 compilers that do not support
2772 variable argument macros. Our copy now differs from upstream.
2778 TF/01 Failure to get a lock on a hints database can have serious
2779 consequences so log it to the panic log.
2781 TF/02 Log LMTP confirmation messages in the same way as SMTP,
2782 controlled using the smtp_confirmation log selector.
2784 TF/03 Include the error message when we fail to unlink a spool file.
2786 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
2787 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
2788 for maintaining out-of-tree patches for some time.
2790 PP/01 Bugzilla 139: Documentation and portability issues.
2791 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
2792 Handle per-OS dynamic-module compilation flags.
2794 PP/02 Let /dev/null have normal permissions.
2795 The 4.73 fixes were a little too stringent and complained about the
2796 permissions on /dev/null. Exempt it from some checks.
2797 Reported by Andreas M. Kirchwitz.
2799 PP/03 Report version information for many libraries, including
2800 Exim version information for dynamically loaded libraries. Created
2801 version.h, now support a version extension string for distributors
2802 who patch heavily. Dynamic module ABI change.
2804 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
2805 privilege escalation vulnerability whereby the Exim run-time user
2806 can cause root to append content of the attacker's choosing to
2809 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
2812 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
2813 If dropping privileges for untrusted macros, we disabled normal logging
2814 on the basis that it would fail; for the Exim run-time user, this is not
2815 the case, and it resulted in successful deliveries going unlogged.
2816 Fixed. Reported by Andreas Metzler.
2822 PP/01 Date: & Message-Id: revert to normally being appended to a message,
2823 only prepend for the Resent-* case. Fixes regression introduced in
2824 Exim 4.70 by NM/22 for Bugzilla 607.
2826 PP/02 Include check_rfc2047_length in configure.default because we're seeing
2827 increasing numbers of administrators be bitten by this.
2829 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
2831 PP/03 Bugzilla 994: added openssl_options main configuration option.
2833 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
2835 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
2837 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
2839 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
2840 without a peer certificate, leading to a segfault because of an
2841 assumption that peers always have certificates. Be a little more
2842 paranoid. Problem reported by Martin Tscholak.
2844 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
2845 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
2846 NB: ClamAV planning to remove STREAM in "middle of 2010".
2847 CL also introduces -bmalware, various -d+acl logging additions and
2848 more caution in buffer sizes.
2850 PP/09 Implemented reverse_ip expansion operator.
2852 PP/10 Bugzilla 937: provide a "debug" ACL control.
2854 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
2856 PP/12 Bugzilla 973: Implement --version.
2858 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
2860 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
2862 PP/15 Bugzilla 816: support multiple condition rules on Routers.
2864 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
2865 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
2866 ignore trailing whitespace.
2868 JJ/02 prevent non-panic DKIM error from being sent to paniclog
2870 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
2873 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
2874 Notification from Dr Andrew Aitchison.
2876 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
2877 ExtendedDetectionInfo response format.
2878 Notification from John Horne.
2880 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
2883 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
2884 XSL and documented dependency on system catalogs, with examples of how
2887 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
2890 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
2891 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
2892 configuration file which is writeable by the Exim user or group.
2894 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
2895 of configuration files to cover files specified with the -C option if
2896 they are going to be used with root privileges, not just the default
2899 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
2900 option (effectively making it always true).
2902 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
2903 files to be used while preserving root privileges.
2905 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
2906 that rogue child processes cannot use them.
2908 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
2909 run-time user, instead of root.
2911 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
2912 Exim run-time user without dropping privileges.
2914 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
2915 result string, instead of calling string_vformat() twice with the same
2918 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
2919 for other users. Others should always drop root privileges if they use
2920 -C on the command line, even for a whitelisted configure file.
2922 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
2924 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
2930 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
2931 $data_path, and $header_path variables; fixed documentation bugs and
2934 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
2935 exipick to access non-standard spools, including the "frozen" queue
2938 NM/01 Bugzilla 965: Support mysql stored procedures.
2939 Patch from Alain Williams
2941 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
2943 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
2944 Patch from Andreas Metzler
2946 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
2947 Patch from Kirill Miazine
2949 NM/05 Bugzilla 671: Added umask to procmail example.
2951 JJ/03 installed exipick 20100323.0, fixing doc bug
2953 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
2954 directory. Notification and patch from Dan Rosenberg.
2956 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
2958 TK/02 Improve log output when DKIM signing operation fails.
2960 MH/01 Treat the transport option dkim_domain as a colon separated
2961 list, not as a single string, and sign the message with each element,
2962 omitting multiple occurences of the same signer.
2964 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
2965 Bugzilla 985, 986. Patch by Simon Arlott
2967 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
2968 Patch by Simon Arlott
2970 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
2971 MBX locking. Notification from Dan Rosenberg.
2977 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
2979 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
2981 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
2983 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
2985 NM/04 Bugzilla 915: Flag broken perl installation during build.
2991 TK/01 Added patch by Johannes Berg that expands the main option
2992 "spamd_address" if it starts with a dollar sign.
2994 TK/02 Write list of recipients to X-Envelope-Sender header when building
2995 the mbox-format spool file for content scanning (suggested by Jakob
2998 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
2999 (http://www.dcc-servers.net/) support via dccifd. Activated by
3000 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
3002 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
3003 by Mark Daniel Reidel <mr@df.eu>.
3005 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
3006 When building exim an external PCRE library is now needed -
3007 PCRE is a system library on the majority of modern systems.
3008 See entry on PCRE_LIBS in EDITME file.
3010 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
3011 conversation. Added nologin parameter to request.
3012 Patch contributed by Kirill Miazine.
3014 TF/01 Do not log submission mode rewrites if they do not change the address.
3016 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
3018 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
3019 log files in place. Contributed by Roberto Lima.
3021 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
3023 TF/03 Bugzilla 615: When checking the local_parts router precondition
3024 after a local_part_suffix or local_part_prefix option, Exim now
3025 does not use the address's named list lookup cache, since this
3026 contains cached lookups for the whole local part.
3028 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
3029 Robert Millan. Documentation is in experimental-spec.txt.
3031 TF/04 Bugzilla 668: Fix parallel build (make -j).
3033 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
3035 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
3036 Patch provided by Jan Srzednicki.
3038 TF/05 Leading white space used to be stripped from $spam_report which
3039 wrecked the formatting. Now it is preserved.
3041 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
3042 that they are available at delivery time.
3044 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
3046 TF/08 TLS error reporting now respects the incoming_interface and
3047 incoming_port log selectors.
3049 TF/09 Produce a more useful error message if an SMTP transport's hosts
3050 setting expands to an empty string.
3052 NM/06 Bugzilla 744: EXPN did not work under TLS.
3053 Patch provided by Phil Pennock.
3055 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
3056 Patch provided by Richard Godbee.
3058 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
3059 acl_smtp_notquit, added index entry.
3061 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
3062 Patch provided by Eugene Bujak.
3064 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
3065 accept(). Patch provided by Maxim Dounin.
3067 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
3068 Patch provided by Phil Pennock.
3070 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
3072 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
3073 Patch provided by Brad "anomie" Jorsch.
3075 NM/14 Bugzilla 622: Added timeout setting to transport filter.
3076 Patch provided by Dean Brooks.
3078 TK/05 Add native DKIM support (does not depend on external libraries).
3080 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
3081 Patch provided by Graeme Fowler.
3083 NM/16 Bugzilla 851: Documentation example syntax fix.
3085 NM/17 Changed NOTICE file to remove references to embedded PCRE.
3087 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
3090 NM/19 Bugzilla 745: TLS version reporting.
3091 Patch provided by Phil Pennock.
3093 NM/20 Bugzilla 167: bool: condition support.
3094 Patch provided by Phil Pennock.
3096 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
3097 clients. Patch provided by Phil Pennock.
3099 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
3100 Patch provided by Brad "anomie" Jorsch.
3102 NM/23 Bugzilla 687: Fix misparses in eximstats.
3103 Patch provided by Heiko Schlittermann.
3105 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
3106 Patch provided by Heiko Schlittermann.
3108 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
3109 plus update to original patch.
3111 NM/26 Bugzilla 799: Documentation correction for ratelimit.
3113 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
3114 Patch provided by David Brownlee.
3116 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
3118 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
3120 NM/30 Bugzilla 888: TLS documentation bugfixes.
3122 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
3124 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
3125 Unlike the original bugzilla I have changed all shell scripts in src tree.
3127 NM/33 Bugzilla 898: Transport filter timeout fix.
3128 Patch by Todd Rinaldo.
3130 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
3131 Patch by Serge Demonchaux.
3133 NM/35 Bugzilla 39: Base64 decode bug fixes.
3134 Patch by Jakob Hirsch.
3136 NM/36 Bugzilla 909: Correct connect() call in dcc code.
3138 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
3140 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
3142 NM/39 Bugzilla 911: Fixed MakeLinks build script.
3148 TK/01 Add preliminary DKIM support. Currently requires a forked version of
3149 ALT-N's libdkim that I have put here:
3150 http://duncanthrax.net/exim-experimental/
3152 Note to Michael Haardt: I had to rename some vars in sieve.c. They
3153 were called 'true' and it seems that C99 defines that as a reserved
3154 keyword to be used with 'bool' variable types. That means you could
3155 not include C99-style headers which use bools without triggering
3156 build errors in sieve.c.
3158 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
3159 as mailq or other aliases. Changed the --help handling significantly
3160 to do whats expected. exim_usage() emits usage/help information.
3162 SC/01 Added the -bylocaldomain option to eximstats.
3164 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
3166 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
3168 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
3174 PH/01 Another patch from the Sieve maintainer.
3176 PH/02 When an IPv6 address is converted to a string for single-key lookup
3177 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
3178 dots are used instead of colons so that keys in lsearch files need not
3179 contain colons. This was done some time before quoting was made available
3180 in lsearch files. However, iplsearch files do require colons in IPv6 keys
3181 (notated using the quote facility) so as to distinguish them from IPv4
3182 keys. This meant that lookups for IP addresses in host lists did not work
3183 for iplsearch lookups.
3185 This has been fixed by arranging for IPv6 addresses to be expressed with
3186 colons if the lookup type is iplsearch. This is not incompatible, because
3187 previously such lookups could never work.
3189 The situation is now rather anomalous, since one *can* have colons in
3190 ordinary lsearch keys. However, making the change in all cases is
3191 incompatible and would probably break a number of configurations.
3193 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
3196 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
3197 conversion specification without a maximum field width, thereby enabling
3198 a rogue spamd server to cause a buffer overflow. While nobody in their
3199 right mind would setup Exim to query an untrusted spamd server, an
3200 attacker that gains access to a server running spamd could potentially
3201 exploit this vulnerability to run arbitrary code as the Exim user.
3203 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
3204 $primary_hostname instead of what libspf2 thinks the hosts name is.
3206 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
3207 a directory entry by the name of the lookup key. Previously, if a
3208 symlink pointed to a non-existing file or a file in a directory that
3209 Exim lacked permissions to read, a lookup for a key matching that
3210 symlink would fail. Now it is enough that a matching directory entry
3211 exists, symlink or not. (Bugzilla 503.)
3213 PH/03 The body_linecount and body_zerocount variables are now exported in the
3216 PH/04 Added the $dnslist_matched variable.
3218 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
3219 This means they are set thereafter only if the connection becomes
3222 PH/06 Added the client_condition to authenticators so that some can be skipped
3223 by clients under certain conditions.
3225 PH/07 The error message for a badly-placed control=no_multiline_responses left
3226 "_responses" off the end of the name.
3228 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
3230 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
3231 (without spaces) instead of just copying the configuration text.
3233 PH/10 Added the /noupdate option to the ratelimit ACL condition.
3235 PH/11 Added $max_received_linelength.
3237 PH/12 Added +ignore_defer and +include_defer to host lists.
3239 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
3240 way in which PCRE > 7.0 is built.
3242 PH/14 Implemented queue_only_load_latch.
3244 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
3245 MAIL command. The effect was to mangle the value on 64-bit systems.
3247 PH/16 Another patch from the Sieve maintainer.
3249 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
3251 PH/18 If a system quota error occurred while trying to create the file for
3252 a maildir delivery, the message "Mailbox is full" was not appended to the
3253 bounce if the delivery eventually timed out. Change 4.67/27 below applied
3254 only to a quota excession during the actual writing of the file.
3256 PH/19 It seems that peer DN values may contain newlines (and other non-printing
3257 characters?) which causes problems in log lines. The DN values are now
3258 passed through string_printing() before being added to log lines.
3260 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
3261 and InterBase are left for another time.)
3263 PH/21 Added message_body_newlines option.
3265 PH/22 Guard against possible overflow in moan_check_errorcopy().
3267 PH/23 POSIX allows open() to be a macro; guard against that.
3269 PH/24 If the recipient of an error message contained an @ in the local part
3270 (suitably quoted, of course), incorrect values were put in $domain and
3271 $local_part during the evaluation of errors_copy.
3277 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
3278 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
3281 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
3282 issue a MAIL command.
3284 PH/02 In an ACL statement such as
3286 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
3288 if a client was not listed at all, or was listed with a value other than
3289 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
3290 the condition was not true (as it should be), so access was not denied.
3291 The bug was that the ! inversion was incorrectly passed on to the second
3292 item. This has been fixed.
3294 PH/03 Added additional dnslists conditions == and =& which are different from
3295 = and & when the dns lookup returns more than one IP address.
3297 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
3298 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
3300 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
3301 FSYNC, which compiles an option called disable_fsync that allows for
3302 bypassing fsync(). The documentation is heavily laced with warnings.
3304 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
3306 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
3307 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
3308 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
3309 including adding "make clean"; (3) Added -fPIC when compiling the test
3310 dynamically loaded module, to get rid of a warning.
3312 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
3313 message fails, move_frozen_messages = true and ignore_bounce_errors_after
3314 = 0s. The bug is otherwise harmless.
3316 PH/07 There was a bug in the dovecot authenticator such that the value of
3317 $auth1 could be overwritten, and so not correctly preserved, after a
3318 successful authentication. This usually meant that the value preserved by
3319 the server_setid option was incorrect.
3321 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
3323 PH/09 Installed PCRE release 7.0.
3325 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
3326 run for batched SMTP input. It is now run at the start of every message
3327 in the batch. While fixing this I discovered that the process information
3328 (output by running exiwhat) was not always getting set for -bs and -bS
3329 input. This is fixed, and it now also says "batched" for BSMTP.
3331 PH/11 Added control=no_pipelining.
3333 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
3334 patch, slightly modified), and move the expansion of helo_data till after
3335 the connection is made in the smtp transport (so it can use these
3338 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
3340 PH/14 Added log_selector = +pid.
3342 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
3344 PH/16 Add ${if forany and ${if forall.
3346 PH/17 Added dsn_from option to vary the From: line in DSNs.
3348 PH/18 Flush SMTP output before performing a callout, unless control =
3349 no_callout_flush is set.
3351 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
3352 was true (the default) a successful delivery failed to delete the retry
3353 item, thus causing premature timeout of the address. The bug is now
3356 PH/20 Added hosts_avoid_pipelining to the smtp transport.
3358 PH/21 Long custom messages for fakedefer and fakereject are now split up
3359 into multiline responses in the same way that messages for "deny" and
3360 other ACL rejections are.
3362 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
3363 with slight modification.
3365 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
3366 draft, changing the syntax and factoring some duplicate code".
3368 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
3369 for deliveries of the second and subsequent messages over the same SMTP
3372 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
3373 ${reduce, with only minor "tidies".
3375 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
3377 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
3378 expansion side effects.
3380 PH/27 When a message times out after an over-quota error from an Exim-imposed
3381 quota, the bounce message says "mailbox is full". This message was not
3382 being given when it was a system quota that was exceeded. It now should
3385 MH/03 Made $recipients available in local_scan(). local_scan() already has
3386 better access to the recipient list through recipients_list[], but
3387 $recipients can be useful in postmaster-provided expansion strings.
3389 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
3390 in the case of a MAIL command with additional options following the
3391 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
3392 were accidentally chopped off.
3394 PH/29 SMTP synchronization checks are implemented when a command is read -
3395 there is a check that no more input is waiting when there shouldn't be
3396 any. However, for some commands, a delay in an ACL can mean that it is
3397 some time before the response is written. In this time, more input might
3398 arrive, invalidly. So now there are extra checks after an ACL has run for
3399 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
3400 pipelining has not been advertised.
3402 PH/30 MH's patch to allow iscntrl() characters to be list separators.
3404 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
3405 returned in the SMTP response when smtp_return_error_details was false.
3406 This has been fixed.
3408 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
3409 instead of the C I/O that was originally supplied, because problems were
3410 reported on Solaris.
3412 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
3413 Exim which did not show up earlier: it was assuming that a call to
3414 SSL_CTX_set_info_callback() might give an error value. In fact, there is
3415 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
3416 was a macro that became an assignment, so it seemed to work. This has
3417 changed to a proper function call with a void return, hence the compile
3418 error. Exim's code has been fixed.
3420 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
3423 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
3425 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
3427 PH/37 If a message is not accepted after it has had an id assigned (e.g.
3428 because it turns out to be too big or there is a timeout) there is no
3429 "Completed" line in the log. When some messages of this type were
3430 selected by exigrep, they were listed as "not completed". Others were
3431 picked up by some special patterns. I have improved the selection
3432 criteria to be more general.
3434 PH/38 The host_find_failed option in the manualroute router can now be set
3435 to "ignore", to completely ignore a host whose IP address cannot be
3436 found. If all hosts are ignored, the behaviour is controlled by the new
3437 host_all_ignored option.
3439 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
3440 homing or because of multiple MX records with /mx) generated more than
3441 one IP address, and the following item turned out to be the local host,
3442 all the secondary addresses of the first item were incorrectly removed
3443 from the list, along with the local host and any following hosts (which
3444 is what is supposed to happen).
3446 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
3447 whoever called Exim into the -H file. In the case of the daemon it was
3448 behaving confusingly. When first started, it used values for whoever
3449 started the daemon, but after a SIGHUP it used the Exim user (because it
3450 calls itself on a restart). I have changed the code so that it now always
3453 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
3454 message are rejected with the same error (e.g. no authentication or bad
3455 sender address), and a DATA command is nevertheless sent (as can happen
3456 with PIPELINING or a stupid MUA), the error message that was given to the
3457 RCPT commands is included in the rejection of the DATA command. This is
3458 intended to be helpful for MUAs that show only the final error to their
3461 PH/42 Another patch from the Sieve maintainer.
3463 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
3464 Eximstats - Fixed some broken HTML links and added missing column headers
3466 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
3467 columns for Rejects, Temp Rejects, Ham, and Spam rows.
3469 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
3471 PH/43 Yet another patch from the Sieve maintainer.
3473 PH/44 I found a way to check for a TCP/IP connection going away before sending
3474 the response to the final '.' that terminates a message, but only in the
3475 case where the client has not sent further data following the '.'
3476 (unfortunately, this is allowed). However, in many cases there won't be
3477 any further data because there won't be any more messages to send. A call
3478 to select() can be used: if it shows that the input is "ready", there is
3479 either input waiting, or the socket has been closed. An attempt to read
3480 the next input character can distinguish the two cases. Previously, Exim
3481 would have sent an OK response which the client would never have see.
3482 This could lead to message repetition. This fix should cure that, at
3483 least in a lot of common cases.
3485 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
3486 advertised in response to EHLO.
3492 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
3493 fixed by 4.65/MH/01 (is this a record?) are fixed:
3495 (i) An empty string was always treated as zero by the numeric comparison
3496 operators. This behaviour has been restored.
3498 (ii) It is documented that the numeric comparison operators always treat
3499 their arguments as decimal numbers. This was broken in that numbers
3500 starting with 0 were being interpreted as octal.
3502 While fixing these problems I realized that there was another issue that
3503 hadn't been noticed. Values of message_size_limit (both the global option
3504 and the transport option) were treated as octal if they started with 0.
3505 The documentation was vague. These values are now always treated as
3506 decimal, and I will make that clear in the documentation.
3512 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
3513 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
3516 MH/01 Don't check that the operands of numeric comparison operators are
3517 integers when their expansion is in "skipping" mode (fixes bug
3518 introduced by 4.64-PH/07).
3520 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
3521 child addresses, Exim now panics and dies. Previously, because the count
3522 is held in a short int, deliveries were likely to be lost. As such a
3523 large number of recipients for a single message is ridiculous
3524 (performance will be very, very poor), I have chosen to impose a limit
3525 rather than extend the field.
3531 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
3532 leftover -K file (the existence of which was triggered by #402).
3533 While we were at it, introduced process PID as part of the -K
3534 filename. This should rule out race conditions when creating
3537 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
3538 processing considerably. Previous code took too long for large mails,
3539 triggering a timeout which in turn triggers #401.
3541 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
3542 in the DK code in transports.c. sendfile() is not really portable,
3543 hence the _LINUX specificness.
3545 TF/01 In the add_headers option to the mail command in an Exim filter,
3546 there was a bug that Exim would claim a syntax error in any
3547 header after the first one which had an odd number of characters
3550 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
3551 callout verification, Exim cached a "reject" for the entire domain. This
3552 is correct for most verifications, but it is not correct for a recipient
3553 verification with use_sender or use_postmaster set, because in that case
3554 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
3555 case of MAIL FROM:<> rejection from other early rejections (e.g.
3556 rejection of HELO). When verifying a recipient using a non-null MAIL
3557 address, the cache is ignored if it shows MAIL FROM:<> rejection.
3558 Whatever the result of the callout, the value of the domain cache is
3559 left unchanged (for any other kind of callout, getting as far as trying
3560 RCPT means that the domain itself is ok).
3562 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
3565 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
3566 manifest itself as EPIPE rather than ECONNECT. When tidying away a
3567 session, the daemon ignores ECONNECT errors and logs others; it now
3568 ignores EPIPE as well.
3570 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
3571 (quoted-printable decoding).
3573 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
3574 later the small subsequent patch to fix an introduced bug.
3576 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
3578 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
3580 PH/08 An error is now given if message_size_limit is specified negative.
3582 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
3583 to be given (somewhat) arbitrary names.
3585 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
3588 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
3589 miscellaneous code fixes
3591 PH/10 Added the log_reject_target ACL modifier to specify where to log
3594 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
3595 hostname. This is wrong, because it relates to the incoming message (and
3596 probably the interface on which it is arriving) and not to the outgoing
3597 callout (which could be using a different interface). This has been
3598 changed to use the value of the helo_data option from the smtp transport
3599 instead - this is what is used when a message is actually being sent. If
3600 there is no remote transport (possible with a router that sets up host
3601 addresses), $smtp_active_hostname is used.
3603 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
3604 tweaks were necessary in order to get it to work (see also 21 below):
3605 (a) The code assumed that strncpy() returns a negative number on buffer
3606 overflow, which isn't the case. Replaced with Exim's string_format()
3608 (b) There were several signed/unsigned issues. I just did the minimum
3609 hacking in of casts. There is scope for a larger refactoring.
3610 (c) The code used strcasecmp() which is not a standard C function.
3611 Replaced with Exim's strcmpic() function.
3612 (d) The code set only $1; it now sets $auth1 as well.
3613 (e) A simple test gave the error "authentication client didn't specify
3614 service in request". It would seem that Dovecot has changed its
3615 interface. Fortunately there's a specification; I followed it and
3616 changed what the client sends and it appears to be working now.
3618 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
3621 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
3622 address A is aliased to B and C, where B exists and C does not. Without
3623 -v the output is "A verified" because verification stops after a
3624 successful redirection if more than one address is generated. However,
3625 with -v the child addresses are also verified. Exim was outputting "A
3626 failed to verify" and then showing the successful verification for C,
3627 with its parentage. It now outputs "B failed to verify", showing B's
3628 parentage before showing the successful verification of C.
3630 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
3631 look up a TXT record in a specific list after matching in a combined
3634 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
3635 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
3636 they consult the DNS. I had assumed they would set it the way they
3637 wanted; and indeed my experiments on Linux seem to show that in some
3638 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
3639 To be on the safe side, however, I have now made the interface to
3640 host_find_byname() similar to host_find_bydns(), with an argument
3641 containing the DNS resolver options. The host_find_byname() function now
3642 sets these options at its start, just as host_find_bydns() does. The smtp
3643 transport options dns_qualify_single and dns_search_parents are passed to
3644 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
3645 of host_find_byname() use the default settings of RES_DEFNAMES
3646 (qualify_single) but not RES_DNSRCH (search_parents).
3648 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
3649 spool_read_header() do less string testing, by means of a preliminary
3650 switch on the second character of optional "-foo" lines. (This is
3651 overdue, caused by the large number of possibilities that now exist.
3652 Originally there were few.) While I was there, I also converted the
3653 str(n)cmp tests so they don't re-test the leading "-" and the first
3654 character, in the hope this might squeeze out yet more improvement.
3656 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
3657 flag allowing group syntax was set by the header_syntax check but not
3658 turned off, possible causing trouble later; (2) The flag was not being
3659 set at all for the header_verify test, causing "group"-style headers to
3660 be rejected. I have now set it in this case, and also caused header_
3661 verify to ignore an empty address taken from a group. While doing this, I
3662 came across some other cases where the code for allowing group syntax
3663 while scanning a header line wasn't quite right (mostly, not resetting
3664 the flag correctly in the right place). These bugs could have caused
3665 trouble for malformed header lines. I hope it is now all correct.
3667 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
3668 with the "reply" argument non-NULL. The code, however (which originally
3669 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
3670 but it didn't always do it. This confused somebody who was copying the
3671 code for some other use. I have removed all the tests.
3673 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
3674 feature that was used to support insecure browsers during the U.S. crypto
3675 embargo. It requires special client support, and Exim is probably the
3676 only MTA that supported it -- and would never use it because real RSA is
3677 always available. This code has been removed, because it had the bad
3678 effect of slowing Exim down by computing (never used) parameters for the
3679 RSA_EXPORT functionality.
3681 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
3682 authenticator to fail if there's a tab character in the incoming data
3683 (there should never be unless someone is messing about, as it's supposed
3684 to be base64-encoded). Also added, on Timo's advice, the "secured" option
3685 if the connection is using TLS or if the remote IP is the same as the
3686 local IP, and the "valid-client-cert option" if a client certificate has
3689 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
3690 authenticators. This can be used for authorization after authentication
3691 succeeds. (In the case of plaintext, it servers for both authentication
3694 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
3695 if any retry times were supplied.
3697 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
3698 connection, where there is no client IP address to check. In this
3699 situation, the verify now always succeeds.
3701 PH/25 Applied John Jetmore's -Mset patch.
3703 PH/26 Added -bem to be like -Mset, but loading a message from a file.
3705 PH/27 In a string expansion for a processed (not raw) header when multiple
3706 headers of the same name were present, leading whitespace was being
3707 removed from all of them, but trailing whitespace was being removed only
3708 from the last one. Now trailing whitespace is removed from each header
3709 before concatenation. Completely empty headers in a concatenation (as
3710 before) are ignored.
3712 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
3713 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
3715 PH/29 [Removed. This was a change that I later backed out, and forgot to
3716 correct the ChangeLog entry (that I had efficiently created) before
3717 committing the later change.]
3719 PH/30 Exim was sometimes attempting to deliver messages that had suffered
3720 address errors (4xx response to RCPT) over the same connection as other
3721 messages routed to the same hosts. Such deliveries are always "forced",
3722 so retry times are not inspected. This resulted in far too many retries
3723 for the affected addresses. The effect occurred only when there were more
3724 hosts than the hosts_max_try setting in the smtp transport when it had
3725 the 4xx errors. Those hosts that it had tried were not added to the list
3726 of hosts for which the message was waiting, so if all were tried, there
3727 was no problem. Two fixes have been applied:
3729 (i) If there are any address or message errors in an SMTP delivery, none
3730 of the hosts (tried or untried) are now added to the list of hosts
3731 for which the message is waiting, so the message should not be a
3732 candidate for sending over the same connection that was used for a
3733 successful delivery of some other message. This seems entirely
3734 reasonable: after all the message is NOT "waiting for some host".
3735 This is so "obvious" that I'm not sure why it wasn't done
3736 previously. Hope I haven't missed anything, but it can't do any
3737 harm, as the worst effect is to miss an optimization.
3739 (ii) If, despite (i), such a delivery is accidentally attempted, the
3740 routing retry time is respected, so at least it doesn't keep
3741 hammering the server.
3743 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
3744 in ${readsocket because some servers need this prod.
3746 PH/32 Added some extra debug output when updating a wait-xxx database.
3748 PH/33 The hint "could be header name not terminated by colon", which has been
3749 given for certain expansion errors for a long time, was not being given
3750 for the ${if def:h_colon_omitted{... case.
3752 PH/34 The spec says: "With one important exception, whenever a domain list is
3753 being scanned, $domain contains the subject domain." There was at least
3754 one case where this was not true.
3756 PH/35 The error "getsockname() failed: connection reset by peer" was being
3757 written to the panic log as well as the main log, but it isn't really
3758 panic-worthy as it just means the connection died rather early on. I have
3759 removed the panic log writing for the ECONNRESET error when getsockname()
3762 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
3763 runs only) independently of the message's sender address. This meant
3764 that, if the 4xx error was in fact related to the sender, a different
3765 message to the same recipient with a different sender could confuse
3766 things. In particular, this can happen when sending to a greylisting
3767 server, but other circumstances could also provoke similar problems.
3768 I have changed the default so that the retry time for these errors is now
3769 based a combination of the sender and recipient addresses. This change
3770 can be overridden by setting address_retry_include_sender=false in the
3773 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
3774 remote server are returned as part of bounce messages. This was not
3775 happening for LMTP over a pipe (the lmtp transport), but now it is the
3776 same for both kinds of LMTP.
3778 PH/38 Despite being documented as not happening, Exim was rewriting addresses
3779 in header lines that were in fact CNAMEs. This is no longer the case.
3781 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
3782 and queue runs started by the daemon processed all messages. This has
3783 been fixed so that -R and -S can now usefully be given with -q<time>.
3785 PH/40 Import PCRE release 6.7 (fixes some bugs).
3787 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
3789 PH/42 Give an error if -q is specified more than once.
3791 PH/43 Renamed the variables $interface_address and $interface_port as
3792 $received_ip_address and $received_port, to make it clear that these
3793 values apply to message reception, and not to the outgoing interface when
3794 a message is delivered. (The old names remain recognized, of course.)
3796 PH/44 There was no timeout on the connect() call when using a Unix domain
3797 socket in the ${readsocket expansion. There now is.
3799 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
3800 be meaningful with "accept".
3802 SC/01 Eximstats V1.43
3803 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
3805 SC/02 Eximstats V1.44
3806 Use a glob alias rather than an array ref in the generated
3807 parser. This improves both readability and performance.
3809 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
3810 Collect SpamAssassin and rejection statistics.
3811 Don't display local sender or destination tables unless
3812 there is data to show.
3813 Added average volumes into the top table text output.
3815 SC/04 Eximstats V1.46
3816 Collect data on the number of addresses (recipients)
3817 as well as the number of messages.
3819 SC/05 Eximstats V1.47
3820 Added 'Message too big' to the list of mail rejection
3821 reasons (thanks to Marco Gaiarin).
3823 SC/06 Eximstats V1.48
3824 Mainlog lines which have GMT offsets and are too short to
3825 have a flag are now skipped.
3827 SC/07 Eximstats V1.49 (Alain Williams)
3828 Added the -emptyok flag.
3830 SC/08 Eximstats V1.50
3831 Fixes for obtaining the IP address from reject messages.
3833 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
3834 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
3835 whitespace changes from 4.64-PH/27
3837 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
3840 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
3841 are found, allow negative numbers in numeric criteria)
3843 JJ/06 exipick.20061117.2, added new $message_body_missing variable
3845 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
3846 to match changes made in 4.64-PH/43
3848 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
3850 PH/47 Put in an explicit test for a DNS lookup of an address record where the
3851 "domain" is actually an IP address, and force a failure. This locks out
3852 those revolvers/nameservers that support "A-for-A" lookups, in
3853 contravention of the specifications.
3855 PH/48 When a host name was looked up from an IP address, and the subsequent
3856 forward lookup of the name timed out, the host name was left in
3857 $sender_host_name, contrary to the specification.
3859 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
3860 restricted to single-key lookups, Exim was not diagnosing an error if
3861 * or *@ was used with a query-style lookup.
3863 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
3865 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
3866 long ago, but noone interested enough thought of it. Let's just say that
3867 the "1.1" means that there are some new functions that weren't there at
3868 some point in the past.
3870 PH/51 Error processing for expansion failure of helo_data from an smtp
3871 transport during callout processing was broken.
3873 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
3874 tested/used via the -bh/-bhc/-bs options.
3876 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
3877 bug, fixed in subsequent PCRE releases).
3879 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
3880 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
3882 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
3888 SC/01 Use a glob alias rather than an array ref in eximstats generated
3889 parser. This improves both readability and performance.
3891 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
3892 Don't display local sender or destination tables in eximstats unless
3893 there is data to show.
3894 Added average volumes into the eximstats top table text output.
3896 SC/03 Collect data on the number of addresses (recipients) as well
3897 as the number of messages in eximstats.
3899 TF/01 Correct an error in the documentation for the redirect router. Exim
3900 does (usually) call initgroups() when daemonizing.
3902 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
3903 with consistent privilege compared to when running as a daemon.
3905 TF/03 Note in the spec that $authenticated_id is not set for local
3906 submissions from trusted users.
3908 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
3909 Thanks to Dean Brooks <dean@iglou.com> for the patch.
3911 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
3912 by adding some example configuration directives to the default
3913 configuration file. A little bit of work is required to uncomment the
3914 directives and define how usernames and passwords are checked, but
3915 there is now a framework to start from.
3917 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
3918 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
3919 without this. I don't know how relevant this is to other LDAP libraries.
3921 PH/02 Add the verb name to the "unknown ACL verb" error.
3923 PH/03 Magnus Holmgren's patch for filter_prepend_home.
3925 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
3927 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
3928 directory not expanded when it should be if an expanded home directory
3929 was set for the address (which is overridden by the transport).
3931 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
3934 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
3935 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
3936 because it is too late at that time, and has no effect.
3938 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
3939 security issue with \' (bugzilla #107). I could not use the
3940 PQescapeStringConn() function, because it needs a PGconn value as one of
3943 PH/08 When testing addresses using -bt, indicate those final addresses that
3944 are duplicates that would not cause an additional delivery. At least one
3945 person was confused, thinking that -bt output corresponded to deliveries.
3946 (Suppressing duplicates isn't a good idea as you lose the information
3947 about possibly different redirections that led to the duplicates.)
3949 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
3950 systems where poll() doesn't work, in particular OS X.
3952 PH/10 Added more information to debugging output for retry time not reached.
3954 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
3955 operations in malware.c.
3957 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
3960 PH/13 If write_rejectlog was set false when logging was sent to syslog with
3961 syslog_duplication set false, log lines that would normally be written
3962 both the the main log and to the reject log were not written to syslog at
3965 PH/14 In the default configuration, change the use of "message" in ACL warn
3966 statements to "add_header".
3968 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
3969 not followed by a command (e.g. "seen endif").
3971 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
3972 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
3975 PH/17 Added extra conditions to the default value of delay_warning_condition
3979 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
3980 { match{$h_precedence:}{(?i)bulk|list|junk} } \
3981 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
3984 The Auto-Submitted: and various List- headers are standardised, whereas I
3985 don't think Precedence: ever was.
3987 PH/18 Refactored debugging code in route_finduser() to show more information,
3988 in particular, the error code if getpwnam() issues one.
3990 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
3991 This is apparently needed in addition to the PH/07 change above to avoid
3992 any possible encoding problems.
3994 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
3995 but not after initializing Perl.
3997 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
3998 output them only if debugging. By default they are written stderr,
3999 apparently, which is not desirable.
4001 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
4004 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
4007 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
4009 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
4010 authenticated or an ident call has been made. Suppress the default
4011 values for $authenticated_id and $authenticated_sender (but permit -oMai
4012 and -oMas) when testing with -bh.
4014 PH/24 Re-jigged the order of the tests in the default configuration so that the
4015 tests for valid domains and recipients precede the DNS black list and CSA
4016 tests, on the grounds that those ones are more expensive.
4018 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
4019 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
4020 This bug exists in every version of Exim that I still have, right back to
4023 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
4024 However, an attempt to turn on case-sensitivity in a regex key by
4025 including (?-i) didn't work because the subject string was already
4026 lowercased, and the effects were non-intuitive. It turns out that a
4027 one-line patch can be used to allow (?-i) to work as expected.
4033 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
4034 other effects) broke the use of negated acl sub-conditions.
4036 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
4039 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
4040 "Deny" causes Exim to reject the incoming connection with a 554 error.
4041 Unfortunately, if there is a major crisis, such as a disk failure,
4042 tcp-wrappers gives "deny", whereas what one would like would be some
4043 kind of temporary error. A kludge has been added to help with this.
4044 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
4045 554 error is used if errno is still zero or contains ENOENT (which occurs
4046 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
4049 PH/03 Add -lutil to the default FreeBSD LIBS setting.
4051 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
4052 errors. Otherwise a message that provokes a temporary error (when other
4053 messages do not) can cause a whole host to time out.
4055 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
4056 addresses were routed directly to files or pipes from a redirect router.
4057 File deliveries just didn't batch; pipe deliveries might have suffered
4060 PH/06 A failure to get a lock for a hints database would erroneously always say
4061 "Failed to get write lock", even when it was really a read lock.
4063 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
4064 of 0600. This has been changed to use the value of the lockfile_mode
4065 option (which defaults to 0600).
4067 PH/08 Applied small patch from the Sieve maintainer.
4069 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
4070 folder from quota calculations, a direct delivery into this folder messed
4071 up the contents of the maildirsize file. This was because the regex was
4072 used only to exclude .Trash (or whatever) when the size of the mailbox
4073 was calculated. There was no check that a delivery was happening into an
4074 excluded directory. This bug has been fixed by ignoring all quota
4075 processing for deliveries into excluded directories.
4077 PH/10 Added the maildirfolder_create_regex option to appendfile.
4083 PH/01 The code for finding all the local interface addresses on a FreeBSD
4084 system running IPv6 was broken. This may well have applied to all BSD
4085 systems, as well as to others that have similar system calls. The broken
4086 code found IPv4 interfaces correctly, but gave incorrect values for the
4087 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
4088 that it would not match correctly against @[] and not recognize the IPv6
4091 PH/02 The ipliteral router was not recognizing addresses of the form user@
4092 [ipv6:....] because it didn't know about the "ipv6:" prefix.
4094 PH/03 Added disable_ipv6.
4096 PH/04 Changed $reply_address to use the raw form of the headers instead of the
4097 decoded form, because it is most often used to construct To: headers
4098 lines in autoreplies, and the decoded form may well be syntactically
4099 invalid. However, $reply_address has leading white space removed, and all
4100 newlines turned into spaces so that the autoreply transport does not
4103 PH/05 If group was specified without a user on a router, and no group or user
4104 was specified on a transport, the group from the router was ignored.
4106 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
4107 for visible compile-time settings that can be used to change these
4108 numbers, for those that want even more. Backwards compatibility with old
4109 spool files has been maintained. However, going back to a previous Exim
4110 release will lost any variables that are in spool files.
4112 PH/07 Two small changes when running in the test harness: increase delay when
4113 passing a TCP/IP connection to a new process, in case the original
4114 process has to generate a bounce, and remove special handling of
4115 127.0.0.2 (sic), which is no longer necessary.
4117 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
4118 be the same on different OS.
4120 PH/09 Moved a debug statement in filter processing to avoid a race problem when
4123 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
4124 whether --show-vars was specified or not
4126 JJ/02 exipick: Added support for new ACL variable spool format introduced
4129 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
4130 syntactically invalid From: or Reply-to: line, and a filter used this to
4131 generate an autoreply, and therefore failed to obtain an address for the
4132 autoreply, Exim could try to deliver to a non-existent relative file
4133 name, causing unrelated and misleading errors. What now happens is that
4134 it logs this as a hard delivery error, but does not attempt to create a
4137 PH/11 The exinext utility has a -C option for testing purposes, but although
4138 the given file was scanned by exinext itself; it wasn't being passed on
4139 when Exim was called.
4141 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
4142 an end-of-file indication when reading a command response.
4144 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
4145 compiled. In many other places in Exim, IPv6 addresses are always
4146 recognized, so I have changed this. It also means that IPv4 domain
4147 literals of the form [IPV4:n.n.n.n] are now always recognized.
4149 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
4150 used if the router is not running as root, for example, when verifying at
4151 ACL time, or when using -bh. The debugging output from this situation was
4152 non-existent - all you got was a failure to exec. I have made two
4155 (a) Failures to set uid/gid, the current directory, or a process leader
4156 in a subprocess such as that created by queryprogram now generate
4157 suitable debugging output when -d is set.
4159 (b) The queryprogram router detects when it is not running as root,
4160 outputs suitable debugging information if -d is set, and then runs
4161 the subprocess without attempting to change uid/gid.
4163 PH/15 Minor change to Makefile for building test_host (undocumented testing
4166 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
4167 additional section of a DNS packet that returns MX or SRV records.
4168 Instead, it always explicitly searches for A/AAAA records. This avoids
4169 major problems that occur when a DNS server includes only records of one
4170 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
4171 fixed another bug: if SRV records were looked up and the corresponding
4172 address records were *not* found in the additional section, the port
4173 values from the SRV records were lost.
4175 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
4176 using the correct key (the original address) when searching the retry
4177 rules in order to find which one to use for generating the retry hint.
4179 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
4180 adding the default one. Similarly, if it contains a Reply-To: header, the
4181 errors_reply_to option, if set, is not used.
4183 PH/19 When calculating a retry time, Exim used to measure the "time since
4184 failure" by looking at the "first failed" field in the retry record. Now
4185 it does not use this if it is later than than the arrival time of the
4186 message. Instead it uses the arrival time. This makes for better
4187 behaviour in cases where some deliveries succeed, thus re-setting the
4188 "first failed" field. An example is a quota failure for a huge message
4189 when small messages continue to be delivered. Without this change, the
4190 "time since failure" will always be short, possible causing more frequent
4191 delivery attempts for the huge message than are intended.
4192 [Note: This change was subsequently modified - see PH/04 for 4.62.]
4194 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
4195 $1, $2, $3) because the numerical variables can be reset during some
4196 expansion items (e.g. "match"), thereby losing the authentication data.
4198 PH/21 Make -bV show the size of off_t variables so that the test suite can
4199 decide whether to run tests for quotas > 2G.
4201 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
4202 mailbox_size, and mailbox_filecount in the appendfile transport. If a
4203 filecount value is greater than 2G or if a quota value is greater than 2G
4204 on a system where the size of off_t is not greater than 4, a panic error
4207 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
4208 never match. The debug and -bh output now contains an explicit error
4209 message indicating a malformed IPv4 address or mask.
4211 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
4212 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
4213 PH/23 above applies.
4215 PH/25 Do not write to syslog when running in the test harness. The only
4216 occasion when this arises is a failure to open the main or panic logs
4217 (for which there is an explicit test).
4219 PH/26 Added the /no_tell option to "control=freeze".
4221 PH/27 If a host name lookup failed very early in a connection, for example, if
4222 the IP address matched host_lookup and the reverse lookup yielded a name
4223 that did not have a forward lookup, an error message of the form "no IP
4224 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
4225 could be logged. Now it outputs the IP address instead of "NULL".
4227 PH/28 An enabling patch from MH: add new function child_open_exim2() which
4228 allows the sender and the authenticated sender to be set when
4229 submitting a message from within Exim. Since child_open_exim() is
4230 documented for local_scan(), the new function should be too.
4232 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
4233 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
4234 results in an empty string is now treated as unset.
4236 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
4238 PH/31 Added sender_verify_fail logging option.
4240 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
4241 needed by RFC 822 but not by RFC 2822 was commented out. I have now
4242 tidied the source and removed it altogether.
4244 PH/33 When a queue run was abandoned because the load average was too high, a
4245 log line was always written; now it is written only if the queue_run log
4246 selector is set. In addition, the log line for abandonment now contains
4247 information about the queue run such as the pid. This is always present
4248 in "start" and "stop" lines but was omitted from the "abandon" line.
4250 PH/34 Omit spaces between a header name and the colon in the error message that
4251 is given when verify = headers_syntax fails (if there are lots of them,
4252 the message gets confusing).
4254 PH/35 Change the default for dns_check_names_pattern to allow slashes within
4255 names, as there are now some PTR records that contain slashes. This check
4256 is only to protect against broken name servers that fall over on strange
4257 characters, so the fact that it applies to all lookups doesn't matter.
4259 PH/36 Now that the new test suite is complete, we can remove some of the
4260 special code in Exim that was needed for the old test suite. For example,
4261 sorting DNS records because real resolvers return them in an arbitrary
4262 order. The new test suite's fake resolver always returns records in the
4265 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
4266 bounces) except when queue_only is set, to avoid logging races between
4267 the different processes.
4269 PH/38 Panic-die if .include specifies a non-absolute path.
4271 PH/39 A tweak to the "H" retry rule from its user.
4273 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
4274 a label. They prevented compilation on older perls.
4276 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
4277 a warning to be raised on newish perls.
4279 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
4280 on queue. Changes to match documented behaviour of showing count of
4281 messages matching specified criteria.
4283 PH/40 Changed the default ident timeout from 30s to 5s.
4285 PH/41 Added support for the use of login_cap features, on those BSD systems
4286 that have them, for controlling the resources used by pipe deliveries.
4288 PH/42 The content-scanning code uses fopen() to create files in which to put
4289 message data. Previously it was not paying any attention to the mode of
4290 the files. Exim runs with umask(0) because the rest of the code creates
4291 files with open(), and sets the required mode explicitly. Thus, these
4292 files were ending up world-writeable. This was not a big issue, because,
4293 being within the spool directory, they were not world-accessible. I have
4294 created a function called modefopen, which takes an additional mode
4295 argument. It sets umask(777), creates the file, chmods it to the required
4296 mode, then resets the umask. All the relevant calls to fopen() in the
4297 content scanning code have been changed to use this function.
4299 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
4300 to 24 hours. This avoids potential overflow problems when processing G
4301 and H retry rules. I suspect nobody ever tinkers with this value.
4303 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
4305 PH/45 When the plaintext authenticator is running as a client, the server's
4306 challenges are checked to ensure they are valid base64 strings. By
4307 default, the authentication attempt is cancelled if an invalid string is
4308 received. Setting client_ignore_invalid_base64 true ignores these errors.
4309 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
4310 they are received. Thus, the responses can be made to depend on the
4311 challenges. If an invalid string is ignored, an empty string is placed in
4314 PH/46 Messages that are created by the autoreply transport now contains a
4315 References: header, in accordance with RFCs 2822 and 3834.
4317 PH/47 Added authenticated_sender_force to the smtp transport.
4319 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
4321 PH/49 Installed latest patch from the Sieve maintainer.
4323 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
4324 was also set, the appendfile transport was unnecessarily scanning a
4325 directory of message files (e.g. for maildir delivery) to find the count
4326 of files (along with the size), even though it did not need this
4327 information. It now does the scan only if it needs to find either the
4328 size of the count of files.
4330 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
4332 PH/52 Two bugs concerned with error handling when the smtp transport is
4335 (i) Exim was not creating retry information for temporary errors given
4336 for individual recipients after the DATA command when the smtp transport
4337 was used in LMTP mode. This meant that they could be retried too
4338 frequently, and not timed out correctly.
4340 (ii) Exim was setting the flag that allows error details to be returned
4341 for LMTP errors on RCPT commands, but not for LMTP errors for individual
4342 recipients that were returned after the DATA command.
4344 PH/53 This is related to PH/52, but is more general: for any failing address,
4345 when detailed error information was permitted to be returned to the
4346 sender, but the error was temporary, then after the final timeout, only
4347 "retry timeout exceeded" was returned. Now it returns the full error as
4348 well as "retry timeout exceeded".
4350 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
4351 do this, and (what is worse) MTAs that accept it.
4353 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
4354 will now be deprecated.
4356 PH/56 New os.c-cygwin from the Cygwin maintainer.
4358 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
4359 formats (previously only available in exim formats via -bpr, -bpru,
4360 and -bpra. Now also available in native and exiqgrep formats)
4362 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
4363 with very large, slow to parse queues
4365 JJ/08 exipick: added ! as generic prefix to negate any criteria format
4367 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
4369 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
4370 responses to authentication challenges, though it was showing the
4371 challenges; (ii) I've removed the CR characters from the debug output for
4374 PH/58 Allow for the insertion of a newline as well as a space when a string
4375 is turned into more than one encoded-word during RFC 2047 encoding. The
4376 Sieve code now uses this.
4378 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
4379 data_4xx, lost_connection, tls_required.
4381 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
4382 message was being sent as an SMTP response.
4384 PH/61 Add -l and -k options to exicyclog.
4386 PH/62 When verifying, if an address was redirected to one new address, so that
4387 verification continued, and the new address failed or deferred after
4388 having set something in $address_data, the value of $address_data was not
4389 passed back to the ACL. This was different to the case when no
4390 redirection occurred. The value is now passed back in both cases.
4392 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
4393 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
4394 use login_cap.h, so on its own it isn't the distinguishing feature. The
4395 new name refers directly to the setclassresources() function.
4397 PH/65 Added configuration files for NetBSD3.
4399 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
4401 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
4402 is preferred over IPv4.
4404 PH/68 The bounce_return_message and bounce_return_body options were not being
4405 honoured for bounces generated during the reception of non-SMTP messages.
4406 In particular, this applied to messages rejected by the ACL. This bug has
4407 been fixed. However, if bounce_return_message is true and bounce_return_
4408 body is false, the headers that are returned for a non-SMTP message
4409 include only those that have been read before the error was detected.
4410 (In the case of an ACL rejection, they have all been read.)
4412 PH/69 The HTML version of the specification is now built in a directory called
4413 spec_html instead of spec.html, because the latter looks like a path with
4414 a MIME-type, and this confuses some software.
4416 PH/70 Catch two compiler warnings in sieve.c.
4418 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
4419 function verify_get_ident() calls ip_connect() to connect a socket, but
4420 if the "connect()" function timed out, ip_connect() used to close the
4421 socket. However, verify_get_ident() also closes the socket later, and in
4422 between Exim writes to the log, which may get opened at this point. When
4423 the socket was closed in ip_connect(), the log could get the same file
4424 descriptor number as the socket. This naturally causes chaos. The fix is
4425 not to close the socket in ip_connect(); the socket should be closed by
4426 the function that creates it. There was only one place in the code where
4427 this was missing, in the iplookup router, which I don't think anybody now
4428 uses, but I've fixed it anyway.
4430 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
4431 well as to direct DNS lookups. Otherwise the handling of names in host
4432 lists is inconsistent and therefore confusing.
4438 PH/01 Two changes to the default runtime configuration:
4440 (1) Move the checks for relay_from_hosts and authenticated clients from
4441 after to before the (commented out) DNS black list checks.
4443 (2) Add control=submission to the relay_from_hosts and authenticated
4444 clients checks, on the grounds that messages accepted by these
4445 statements are most likely to be submissions.
4447 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
4449 (1) Generate an error if the third argument for the ${prvs expansion is
4452 (2) Treat a missing third argument of ${prvscheck as if it were an empty
4455 (3) Reset the variables that are obtained from the first argument of
4456 ${prvscheck and used in the second argument before leaving the code,
4457 because their memory is reclaimed, so using them afterwards may do
4460 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
4461 one (it's much easier than Tom thought :-).
4463 (5) Because of (4), we can now allow for the use of $prvscheck_result
4464 inside the third argument.
4466 PH/03 For some reason, the default setting of PATH when running a command from
4467 a pipe transport was just "/usr/bin". I have changed it to
4470 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
4471 anything to be listed in the output from -bV.
4473 PH/05 When a filter generated an autoreply, the entire To: header line was
4474 quoted in the delivery log line, like this:
4476 => >A.N.Other <ano@some.domain> <original@ddress> ...
4478 This has been changed so that it extracts the operative address. There
4479 may be more than one such address. If so, they are comma-separated, like
4482 => >ano@some.domain,ona@other.domain <original@ddress> ...
4484 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
4485 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
4486 not being looked up in the rDNS to get a host name, Exim was showing the
4487 IP address twice in Received: lines, even though the IP addresses were
4488 identical. For example:
4490 Received: from [1.2.3.4] (helo=[1.2.3.4])
4492 However, if the real host name was known, it was omitting the HELO data
4493 if it matched the actual IP address. This has been tidied up so that it
4494 doesn't show the same IP address twice.
4496 PH/07 When both +timestamp and +memory debugging was on, the value given by
4497 $tod_xxx expansions could be wrong, because the tod_stamp() function was
4498 called by the debug printing, thereby overwriting the timestamp buffer.
4499 Debugging no longer uses the tod_stamp() function when +timestamp is set.
4501 PH/08 When the original message was included in an autoreply transport, it
4502 always said "this is a copy of the message, including all the headers",
4503 even if body_only or headers_only was set. It now gives an appropriate
4506 PH/09 Applied a patch from the Sieve maintainer which:
4508 o fixes some comments
4509 o adds the (disabled) notify extension core
4510 o adds some debug output for the result of if/elsif tests
4511 o points to the current vacation draft in the documentation
4512 and documents the missing references header update
4516 o fixes a bug in processing the envelope test (when testing
4517 multiple envelope elements, the last element determined the
4520 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
4521 Electronic Mail") by including:
4523 Auto-submitted: auto-generated
4525 in the messages that it generates (bounce messages and others, such as
4526 warnings). In the case of bounce messages for non-SMTP messages, there was
4527 also a typo: it was using "Auto_submitted" (underscore instead of
4528 hyphen). Since every message generated by Exim is necessarily in response
4529 to another message, thes have all been changed to:
4531 Auto-Submitted: auto-replied
4533 in accordance with these statements in the RFC:
4535 The auto-replied keyword:
4537 - SHOULD be used on messages sent in direct response to another
4538 message by an automatic process,
4540 - MUST NOT be used on manually-generated messages,
4542 - MAY be used on Delivery Status Notifications (DSNs) and Message
4543 Disposition Notifications (MDNs),
4545 - MUST NOT be used on messages generated by automatic or periodic
4546 processes, except for messages which are automatic responses to
4549 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
4550 to the default Received: header definition.
4552 PH/12 Added log selector acl_warn_skipped (default on).
4554 PH/13 After a successful wildlsearch lookup, discard the values of numeric
4555 variables because (a) they are in the wrong storage pool and (b) even if
4556 they were copied, it wouldn't work properly because of the caching.
4558 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
4559 checking when decoding. Apparently there are clients that generate
4560 overlong encoded strings. Why am I not surprised?
4562 PH/15 If the first argument of "${if match_address" was not empty, but did not
4563 contain an "@" character, Exim crashed. Now it writes a panic log message
4564 and treats the condition as false.
4566 PH/16 In autoreply, treat an empty string for "once" the same as unset.
4568 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
4569 extension "envelope-auth". The code is finished and in agreement with
4570 other implementations, but there is no documentation so far and in fact,
4571 nobody wrote the draft yet. This extension is currently #undef'ed, thus
4572 not changing the active code.
4574 Print executed "if" and "elsif" statements when debugging is used. This
4575 helps a great deal to understand what a filter does.
4577 Document more things not specified clearly in RFC3028. I had all this
4578 sorted out, when out of a sudden new issues came to my mind. Oops."
4580 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
4583 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
4584 canonical form (as documented). However, after a host name lookup from
4585 the IP address, check_host() was doing a simple string comparison with
4586 addresses acquired from the DNS when checking that the found name did
4587 have the original IP as one of its addresses. Since any found IPv6
4588 addresses are likely to be in abbreviated form, the comparison could
4589 fail. Luckily, there already exists a function for doing the comparison
4590 by converting both addresses to binary, so now that is used instead of
4591 the text comparison.
4593 PH/20 There was another similar case to PH/19, when a complete host name was
4594 given in a host list; looking up its IP address could give an abbreviated
4595 form, whereas the current host's name might or might not be abbreviated.
4596 The same fix has been applied.
4602 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
4603 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
4606 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
4607 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
4609 PH/03 Typo: missing ".o" in src/pcre/Makefile.
4611 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
4612 header line, restrict the check to what is listed in RFCs 2369 and 2929.
4613 Also, for "Auto-Submitted", treat anything other than "no" as
4614 non-personal, in accordance with RFC 3834. (Previously it treated
4615 anything starting "auto-" as non-personal.)
4617 TF/01 The control=submission/name=... option had a problem with syntax
4618 errors if the name included a slash character. The /name= option
4619 now slurps the rest of the string, so it can include any characters
4620 but it must come last in the list of options (after /sender_retain
4623 PH/05 Some modifications to the interface to the fake nameserver for the new
4631 TK/01 Added the "success_on_redirect" address verification option. See
4632 NewStuff for rationale and an example.
4634 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
4636 PH/02 Patch to exigrep to allow it to work on syslog lines.
4638 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
4639 fread() to skip over the body file's header line, because in Cygwin the
4640 header line is locked and is inaccessible.
4642 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
4643 co-exist for some time) to make it clear that it is the Exim ID that is
4644 referenced, not the Message-ID: header line.
4646 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
4647 string_format() function, because snprintf() does not exist on all
4650 PH/06 The use of forbid_filter_existstest now also locks out the use of the
4651 ${stat: expansion item.
4653 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
4654 protocol synchronization error", to keep the pedants happy.
4656 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
4657 well as for IRIX systems, when gcc is being used. See the host.c source
4660 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
4662 PH/10 Named domain lists were not working if used in a queue_smtp_domains
4665 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
4666 transport and to the smtp transport in LMTP mode.
4668 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
4670 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
4671 run a filter in a subprocess. This could lead to confusion in subsequent
4672 lookups in the parent process. There should also be a search_tidyup() at
4673 the end of the subprocess.
4675 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
4676 only if the host matched helo_try_verify_hosts, which caused the
4677 verification to occur when the EHLO/HELO command was issued. The ACL just
4678 tested the remembered result. Now, if a previous verification attempt has
4679 not happened, "verify = helo" does it there and then.
4681 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
4683 TK/03 Fix log output including CR from clamd.
4685 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
4686 exist provoked a memory error which could cause a segfault.
4688 PH/15 Installed PCRE 6.2
4690 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
4692 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
4693 of the problem. Specifically, suggested +O2 rather than +O1 for the
4696 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
4698 PH/20 If a delivery was routed to a non-standard port by means of an SRV
4699 record, the port was not correctly logged when the outgoing_port log
4700 selector was set (it logged the transort's default port).
4702 PH/21 Added support for host-specific ports to manualroute, queryprogram,
4703 fallback_hosts, and "hosts" in the smtp transport.
4705 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
4706 host errors such as "Connection refused".
4708 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
4709 authentication with radiusclient 0.4.9:
4711 - Error returned from rc_read_config was caught wrongly
4712 - Username/password not passed on to radius server due to wrong length.
4714 The presumption is that some radiusclient API changes for 4.51/PH/17
4715 were not taken care of correctly. The code is still untested by me (my
4716 Linux distribution still has 0.3.2 of radiusclient), but it was
4717 contributed by a Radius user.
4719 PH/24 When doing a callout, the value of $domain wasn't set correctly when
4720 expanding the "port" option of the smtp transport.
4722 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
4723 while reading a MIME header. Thanks to Tom Hughes for a patch.
4725 PH/24 Include config.h inside local_scan.h so that configuration settings are
4728 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
4729 that in an ACL for RCPT (for example), you can examine exactly what was
4732 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
4733 commands, but it was not correctly comparing the address with the actual
4734 client host address. Thus, it would show the EHLO address in Received:
4735 header lines when this was not necessary.
4737 PH/27 Added the % operator to ${eval:}.
4739 PH/28 Exim tries to create and chdir to its spool directory when it starts;
4740 it should be ignoring failures (because with -C, for example, it has lost
4741 privilege). It wasn't ignoring creation failures other than "already
4744 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
4747 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
4748 because an input line was too long, either on its own, or by virtue of
4749 too many continuations, the temporary file was not being removed, and the
4750 return code was incorrect.
4752 PH/31 Missing "BOOL" in function definition in filtertest.c.
4754 PH/32 Applied Sieve patches from the maintainer.
4756 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
4758 PH/33 Added "verify = not_blind".
4760 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
4761 Local/Makefile (with some defaults set). These are used in built scripts
4762 such as exicyclog, but they have never been used in the exim_install
4763 script (though there are many overriding facilities there). I have
4764 arranged that the exim_install script now takes note of these two
4767 PH/35 Installed configuration files for Dragonfly.
4769 PH/36 When a locally submitted message by a trusted user did not contain a
4770 From: header, and the sender address was obtained from -f or from an SMTP
4771 MAIL command, and the trusted user did not use -F to supply a sender
4772 name, $originator_name was incorrectly used when constructing a From:
4773 header. Furthermore, $originator_name was used for submission mode
4774 messages from external hosts without From: headers in a similar way,
4775 which is clearly wrong.
4777 PH/37 Added control=suppress_local_fixups.
4779 PH/38 When log_selector = +received_sender was set, and the addition of the
4780 sender made the log line's construction buffer exactly full, or one byte
4781 less than full, an overflow happened when the terminating "\n" was
4784 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
4785 when the result of a list match is failure because a DNS lookup failed.
4787 PH/40 RM_COMMAND is now used in the building process.
4789 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
4790 the "build-* directories that it finds.
4792 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
4793 address in a domain literal was a prefix of an interface address.
4795 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
4796 when verifying a sender address, unless rewrite_headers is false.
4798 PH/44 Wrote a long comment about why errors_to addresses are verified as
4799 recipients, not senders.
4801 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
4802 the ratelimit ACL was added.
4804 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
4806 PH/46 Added extra information about PostgreSQL errors to the error string.
4808 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
4809 avoiding the need to install special zones in a real server. This is
4810 backwards compatible; if it can't find the fake resolver, it drops back.
4811 Thus, both old and new test suites can be run.
4813 TF/02 Added util/ratelimit.pl
4815 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
4816 clock is set back in time.
4818 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
4819 Candler <B.Candler@pobox.com>.
4821 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
4822 OK for addresses that are the result of redirections.
4824 PH/48 A number of further additions for the benefit of the new test suite,
4825 including a fake gethostbyname() that interfaces to the fake DNS resolver
4828 TF/06 The fix for widen_domains has also been applied to qualify_single and
4829 search_parents which are the other dnslookup options that can cause
4832 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
4835 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
4837 TF/07 Exim produced the error message "an SRV record indicated no SMTP
4838 service" if it encountered an MX record with an empty target hostname.
4839 The message is now "an MX or SRV record indicated no SMTP service".
4841 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
4842 if the DNS of the sending site is misconfigured. This is quite a
4843 common situation. This change restores the behaviour of treating a
4844 helo verification defer as a failure.
4846 PH/51 If self=fail was set on a router, the bounce message did not include the
4847 actual error message.
4853 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
4855 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
4856 command itself ended in error, the underlying message about the transport
4857 filter timeout was being overwritten with the pipe command error. Now the
4858 underlying error message should be appended to the second error message.
4860 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
4862 PH/02 Reduce the amount of output that "make" produces by default. Full output
4863 can still be requested.
4865 PH/03 The warning log line about a condition test deferring for a "warn" verb
4866 was being output only once per connection, rather than after each
4867 occurrence (because it was using the same function as for successful
4868 "warn" verbs). This seems wrong, so I have changed it.
4870 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
4871 it should not have, which might have caused a crash in the right
4872 circumstances, but probably never did.
4874 PH/04 Installed a modified version of Tony Finch's patch to make submission
4875 mode fix the return path as well as the Sender: header line, and to
4876 add a /name= option so that you can make the user's friendly name appear
4879 TF/03 Added the control = fakedefer ACL modifier.
4881 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
4882 Mark Lowes for thorough testing.
4884 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
4886 TK/03 Merged latest SRS patch from Miles Wilton.
4888 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
4889 with the definition in sysexits.h (which is #included earlier).
4890 Fortunately, Exim does not actually use EX_OK. The code used to try to
4891 preserve the sysexits.h value, by assuming that macro definitions were
4892 scanned for macro replacements. I have been disabused of this notion,
4893 so now the code just undefines EX_OK before #including unistd.h.
4895 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
4896 in the smtp transport. When a block could not be written in a single
4897 write() function, the timeout was being re-applied to each part-write.
4898 This seems wrong - if the receiver was accepting one byte at a time it
4899 would take for ever. The timeout is now adjusted when this happens. It
4900 doesn't have to be particularly precise.
4902 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
4903 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
4905 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
4906 without a domain if the check to <postmaster@domain> fails.
4908 SC/01 Eximstats: added -xls and the ability to specify output files
4909 (patch written by Frank Heydlauf).
4911 SC/02 Eximstats: use FileHandles for outputting results.
4913 SC/03 Eximstats: allow any combination of xls, txt, and html output.
4915 SC/04 Eximstats: fixed display of large numbers with -nvr option
4917 SC/05 Eximstats: fixed merging of reports with empty tables.
4919 SC/06 Eximstats: added the -include_original_destination flag
4921 SC/07 Eximstats: removed tabs and trailing whitespace.
4923 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
4925 TK/06 MBOX spool code: Add real "From " MBOX separator line
4926 so the .eml file is really in mbox format (even though
4927 most programs do not really care). Patch from Alex Miller.
4929 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
4930 The latter is generated from $received_to and is only set if the
4931 message has one envelope recipient. SA can use these headers,
4932 obviously out-of-the-box. Patch from Alex Miller.
4934 PH/08 The ${def test on a variable was returning false if the variable's
4935 value was "0", contrary to what the specification has always said!
4936 The result should be true unless the variable is empty.
4938 PH/09 The syntax error of a character other than { following "${if
4939 def:variable_name" (after optional whitespace) was not being diagnosed.
4940 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
4941 accidental colon was present, for example, could give incorrect results.
4943 PH/10 Tidied the code in a number of places where the st_size field of a stat()
4944 result is used (not including appendfile, where other changes are about
4947 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
4948 This involved changing a lot of size variables from int to off_t. It
4949 should work with maildirs and everything.
4951 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
4952 spamd dying while we are connected to it.
4954 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
4957 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
4958 function for BDB 4.3.
4960 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
4962 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
4963 the sources for PCRE have been re-arranged and more files are now
4966 PH/15 The code I had for printing potentially long long variables in PH/11
4967 above was not the best (it lost precision). The length of off_t variables
4968 is now inspected at build time, and an appropriate printing format (%ld
4969 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
4970 to be "long long int" or "long int". This is needed for the internal
4971 formatting function string_vformat().
4973 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
4974 the configuration file to be ":syslog", then the script "guesses" where
4975 the logs files are, rather than using the compiled in default. In our
4976 case the guess is not the same as the compiled default, so the script
4977 suddenly stopped working when I started to use syslog. The patch checks
4978 to see if log_file_path is "". If so, it attempts to read it from exim
4979 with no configuration file to get the compiled in version, before it
4980 falls back to the previous guessing code."
4982 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
4983 implementing BATV in an Exim configuration. See NewStuff for the gory
4986 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
4987 Makefile that are specific to HP-UX.
4989 PH/18 If the "use_postmaster" option was set for a recipient callout together
4990 with the "random" option, the postmaster address was used as the MAIL
4991 FROM address for the random test, but not for the subsequent recipient
4992 test. It is now used for both.
4994 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
4995 patch removes a few documentation additions to RFC 3028, because the
4996 latest draft now contains them. It adds the new en;ascii-case comparator
4997 and a new error check for 8bit text in MIME parts. Comparator and
4998 require names are now matched exactly. I enabled the subaddress
4999 extension, but it is not well tested yet (read: it works for me)."
5001 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
5002 rework some of the code of TK/09 above to avoid the hardwired use of
5003 "%lld" and "long long". Replaced the call to snprintf() with a call to
5006 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
5007 records point to non-existent hosts", "retry timeout exceeded", and
5008 "retry time not reached for any host after a long failure period".
5010 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
5011 experimental DomainKeys support:
5013 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
5014 (2) On an error such as an illegally used "control", the wrong name for
5015 the control was given.
5017 These problems did NOT occur unless DomainKeys support was compiled.
5019 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
5021 PH/24 Added ${if match_ip condition.
5023 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
5024 databases so that it will be absolutely obvious if a crash occurs in the
5025 DB library. This is a regular occurrence (often caused by mis-matched
5028 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
5029 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
5030 picked up on a user's system that detects such things. There doesn't seem
5031 to be a gcc warning option for this - only an attribute that has to be
5032 put on the function's prototype. It seems that in Fedora Core 4 they have
5033 set this on a number of new functions. No doubt there will be more in due
5036 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
5037 specify a transport. However, if an address that was verified by such a
5038 router was the subject of a callout, Exim crashed because it tried to
5039 read the rcpt_include_affixes from the non-existent transport. Now it
5040 just assumes that the setting of that option is false. This bug was
5041 introduced by 4.51/PH/31.
5043 PH/28 Changed -d+all to exclude +memory, because that information is very
5044 rarely of interest, but it makes the output a lot bigger. People tend to
5045 do -d+all out of habit.
5047 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
5048 code in os-type was giving problems when libc.so lives in lib64, like on
5051 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
5052 aren't the modern standard, and it seems that some systems' include files
5053 don't always have them. Exim was already checking for some of the newer
5054 ones like T_AAAA, and defining it itself. I've added checks for all the
5055 record types that Exim uses.
5057 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
5058 not automatically generating a new one, as it is supposed to. This
5059 prevented TLS from working. If the file did exist, but contained invalid
5060 data, a new version was generated, as expected. It was only the case of a
5061 non-existent file that was broken.
5063 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
5064 with a change in libdomainkeys > 0.64.
5066 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
5067 from DNS. If the selector record carries the flag, it now has
5068 precedence over the domain-wide flag.
5070 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
5072 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
5073 the use of a port name that isn't defined in /etc/services) occurred, the
5074 message was deferred as in a normal delivery, and thus remained on the
5075 spool, instead of being failed because of the mua_wrapper setting. This
5076 is now fixed, and I tidied up some of the mua_wrapper messages at the
5079 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
5080 the messages in a hash of arrays rather than using individual hashes.
5081 This is a bit cleaner and results in dramatic memory savings, albeit
5082 at a slight CPU cost.
5084 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
5085 as requested by Marc Sherman.
5087 SC/10 Eximstats: added histograms for user specified patterns as requested
5090 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
5092 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
5093 fopen() in the content-scanning modules that did not already have it.
5099 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
5100 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
5102 TK/02 Fix ACL "control" statement not being available in MIME ACL.
5104 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
5106 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
5107 to test Sieve filters that use "vacation".
5109 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
5110 that changes the way the GnuTLS parameters are stored in the cache file.
5111 The new format can be generated externally. For backward compatibility,
5112 if the data in the cache doesn't make sense, Exim assumes it has read an
5113 old-format file, and it generates new data and writes a new file. This
5114 means that you can't go back to an older release without removing the
5117 PH/03 A redirect router that has both "unseen" and "one_time" set does not
5118 work if there are any delivery delays because "one_time" forces the
5119 parent to be marked "delivered", so its unseen clone is never tried
5120 again. For this reason, Exim now forbids the simultaneous setting of
5123 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
5124 redirected to themselves ("homonym" addresses). Read the long ChangeLog
5125 entry if you want to know the details. The fix, however, neglected to
5126 consider the case when local delivery batching is involved. The test for
5127 "previously delivered" was not happening when checking to see if an
5128 address could be batched with a previous (undelivered) one; under
5129 certain circumstances this could lead to multiple deliveries to the same
5132 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
5133 in its include files, and this causes problems building Exim.
5135 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
5136 header_syntax) but Exim was just ignoring anything given after a slash.
5137 In particular, this caused confusion with an attempt to use "verify =
5138 reverse_host_lookup/defer_ok". An error is now given when options are
5139 supplied for verify items that do not have them. (Maybe reverse_host_
5140 lookup should have a defer_ok option, but that's a different point.)
5142 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
5143 defined by RFC 821) to 2048, because there were problems with some AUTH
5144 commands, and RFC 1869 says the size should be increased for extended
5145 SMTP commands that take arguments.
5147 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
5150 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
5151 "unknown" error; now it says that the functionality isn't in the binary.
5153 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
5154 an address' error message when a string expansion fails (syntax or
5155 whatever). Otherwise the password may appear in the log. Following change
5156 PH/42 below, there is no longer a chance of it appearing in a bounce
5159 PH/11 Installed exipick version 20050225.0 from John Jetmore.
5161 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
5162 of its addresses was ever tried. (Bugzilla bug #2.)
5164 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
5165 the result incorrectly in the debug output. (It correctly added a newline
5166 to what was transported.)
5168 TF/01 Added $received_time.
5170 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
5171 commented out examples of how to interface to a virus scanner and to
5172 SpamAssassin. Also added commented examples of av_scanner and
5173 spamd_address settings.
5175 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
5176 and controls are allowed in which ACLs. There were a couple of minor
5177 errors. Some of the entries in the conditions table (which is a table of
5178 where they are NOT allowed) were getting very unwieldy; rewrote them as a
5179 negation of where the condition IS allowed.
5181 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
5183 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
5184 header file does not have a version number, so I've had to invent a new
5185 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
5186 API. The code is untested by me (my Linux distribution still has 0.3.2 of
5187 radiusclient), but it was contributed by a Radius user.
5189 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
5190 files or directories, for OpenSSL.
5192 PH/19 When an Exim process that is running as root has to create an Exim log
5193 file, it does so in a subprocess that runs as exim:exim so as to get the
5194 ownership right at creation (otherwise, other Exim processes might see
5195 the file with the wrong ownership). There was no test for failure of this
5196 fork() call, which would lead to the process getting stuck as it waited
5197 for a non-existent subprocess. Forks do occasionally fail when resources
5198 run out. I reviewed all the other calls to fork(); they all seem to check
5201 PH/20 When checking for unexpected SMTP input at connect time (before writing
5202 the banner), Exim was not dealing correctly with a non-positive return
5203 from the read() function. If the client had disconnected by this time,
5204 the result was a log entry for a synchronization error with an empty
5205 string after "input=" when read() returned zero. If read() returned -1
5206 (an event I could not check), uninitialized data bytes were printed.
5207 There were reports of junk text (parts of files, etc) appearing after
5210 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
5212 PH/22 Added support for macro redefinition, and (re)definition in between
5213 driver and ACL definitions.
5215 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
5216 forgetting to use the resulting value; it was using the unexpanded value.
5218 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
5219 hadn't been configured. The fix is from Juergen Kreileder, who
5220 understands it better than I do:
5222 "Here's what I see happening with three configured cyrus_sasl
5223 authenticators configured (plain, login, cram-md5):
5225 On startup auth_cyrus_sasl_init() gets called for each of these.
5226 This means three calls to sasl_listmech() without a specified mech_list.
5227 => SASL tests which mechs of all available mechs actually work
5228 => three warnings about OTP not working
5229 => the returned list contains: plain, login, cram-md5, digest-md5, ...
5231 With the patch, sasl_listmech() also gets called three times. But now
5232 SASL's mech_list option is set to the server_mech specified in the the
5233 authenticator. Or in other words, the answer from sasl_listmech()
5234 gets limited to just the mech you're testing for (which is different
5236 => the return list contains just 'plain' or 'login', 'cram-md5' or
5237 nothing depending on the value of ob->server_mech.
5239 I've just tested the patch: Authentication still works fine,
5240 unavailable mechs specified in the exim configuration are still
5241 caught, and the auth.log warnings about OTP are gone."
5243 PH/25 When debugging is enabled, the contents of the command line are added
5244 to the debugging output, even when log_selector=+arguments is not
5247 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
5248 answer is "GNU", and only if the return is "GNU/something" is the answer
5251 PH/27 $acl_verify_message is now set immediately after the failure of a
5252 verification in an ACL, and so is available in subsequent modifiers. In
5253 particular, the message can be preserved by coding like this:
5255 warn !verify = sender
5256 set acl_m0 = $acl_verify_message
5258 Previously, $acl_verify_message was set only while expanding "message"
5259 and "log_message" when a very denied access.
5261 PH/28 Modified OS/os.c-Linux with
5263 -#ifndef OS_LOAD_AVERAGE
5264 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
5266 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
5267 nomenclature these days.)
5269 PH/29 Installed patch from the Sieve maintainer that adds the options
5270 sieve_useraddress and sieve_subaddress to the redirect router.
5272 PH/30 In these circumstances:
5273 . Two addresses routed to the same list of hosts;
5274 . First host does not offer TLS;
5275 . First host accepts first address;
5276 . First host gives temporary error to second address;
5277 . Second host offers TLS and a TLS session is established;
5278 . Second host accepts second address.
5279 Exim incorrectly logged both deliveries with the TLS parameters (cipher
5280 and peerdn, if requested) that were in fact used only for the second
5283 PH/31 When doing a callout as part of verifying an address, Exim was not paying
5284 attention to any local part prefix or suffix that was matched by the
5285 router that accepted the address. It now behaves in the same way as it
5286 does for delivery: the affixes are removed from the local part unless
5287 rcpt_include_affixes is set on the transport.
5289 PH/32 Add the sender address, as F=<...>, to the log line when logging a
5290 timeout during the DATA phase of an incoming message.
5292 PH/33 Sieve envelope tests were broken for match types other than :is. I have
5293 applied a patch sanctioned by the Sieve maintainer.
5295 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
5296 the uid or gid is negative. A case of a negative gid caused this to be
5297 noticed. The fix allows for either to be negative.
5299 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
5300 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
5303 PH/36 The change PH/12 above was broken. Fixed it.
5305 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
5306 the grounds that routing the same address twice would always produce the
5307 same answer. This might have been true once, but it is certainly no
5308 longer true now. Routing a child address may depend on the previous
5309 routing that produced that child. Some complicated redirection strategies
5310 went wrong when messages had multiple recipients, and made Exim's
5311 behaviour dependent on the order in which the addresses were given.
5313 I have moved the duplicate checking until after the routing is complete.
5314 Exim scans the addresses that are assigned to local and remote
5315 transports, and removes any duplicates. This means that more work will be
5316 done, as duplicates will always all be routed, but duplicates are
5317 presumably rare, so I don't expect this is of any significance.
5319 For deliveries to pipes, files, and autoreplies, the duplicate checking
5320 still happens during the routing process, since they are not going to be
5323 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
5324 It corrects a timeout issue with spamd. This is Ian's comment: "The
5325 background is that sometimes spamd either never reads data from a
5326 connection it has accepted, or it never writes response data. The exiscan
5327 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
5328 blindly assumes that writes won't block so it may never time out."
5330 PH/39 Allow G after quota size as well as K and M.
5332 PH/40 The value set for $authenticated_id in an authenticator may not contain
5333 binary zeroes or newlines because the value is written to log lines and
5334 to spool files. There was no check on this. Now the value is run through
5335 the string_printing() function so that such characters are converted to
5336 printable escape sequences.
5338 PH/41 $message_linecount is a new variable that contains the total number of
5339 lines in the message. Compare $body_linecount, which is the count for the
5342 PH/42 Exim no longer gives details of delivery errors for specific addresses in
5343 bounce and delay warning messages, except in certain special cases, which
5346 (a) An SMTP error message from a remote host;
5347 (b) A message specified in a :fail: redirection;
5348 (c) A message specified in a "fail" command in a system filter;
5349 (d) A message specified in a FAIL return from the queryprogram router;
5350 (e) A message specified by the cannot_route_message router option.
5352 In these cases only, Exim does include the error details in bounce and
5353 warning messages. There are also a few cases where bland messages such
5354 as "unrouteable address" or "local delivery error" are given.
5356 PH/43 $value is now also set for the "else" part of a ${run expansion.
5358 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
5359 being worked on, but at least Exim now implements the latest version to
5362 PH/45 In a pipe transport, although a timeout while waiting for the pipe
5363 process to complete was treated as a delivery failure, a timeout while
5364 writing the message to the pipe was logged, but erroneously treated as a
5365 successful delivery. Such timeouts include transport filter timeouts. For
5366 consistency with the overall process timeout, these timeouts are now
5367 treated as errors, giving rise to delivery failures by default. However,
5368 there is now a new Boolean option for the pipe transport called
5369 timeout_defer, which, if set TRUE, converts the failures into defers for
5370 both kinds of timeout. A transport filter timeout is now identified in
5373 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
5374 systems where "make" and "gmake" are different, calling "gmake" at top
5375 level broke things. I've arranged for the value of $(MAKE) to be passed
5376 from the Makefile to this script so that it can call the same version of
5380 A note about Exim versions 4.44 and 4.50
5381 ----------------------------------------
5383 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
5384 changes of various kinds. As a consequence, a big documentation update was
5385 needed. This delayed the release for rather longer than seemed good, especially
5386 in the light of a couple of (minor) security issues. Therefore, the changes
5387 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
5388 release. So 4.44 and 4.50 are in effect two different branches that both start
5391 I have left the 4.50 change log unchanged; it contains all the changes since
5392 4.43. The change log for 4.44 is below; many of its items are identical to
5393 those for 4.50. This seems to be the most sensible way to preserve the
5394 historical information.
5400 1. Minor wording change to the doc/README.SIEVE file.
5402 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
5403 computation of the current number of files was incorrect.
5405 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
5406 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
5409 4. Give more explanation in the error message when the command for a transport
5410 filter fails to execute.
5412 5. There are several places where Exim runs a non-Exim command in a
5413 subprocess. The SIGUSR1 signal should be disabled for these processes. This
5414 was being done only for the command run by the queryprogram router. It is
5415 now done for all such subprocesses. The other cases are: ${run, transport
5416 filters, and the commands run by the lmtp and pipe transports.
5418 6. Added CONFIGURE_GROUP build-time option.
5420 7. Some older OS have a limit of 256 on the maximum number of file
5421 descriptors. Exim was using setrlimit() to set 1000 as a large value
5422 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
5423 systems. I've change it so that if it can't get 1000, it tries for 256.
5425 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
5426 was an oversight, and furthermore, ever since the addition of extra
5427 controls (e.g. 4.43/32), the checks on when to allow different forms of
5428 "control" were broken. There should now be diagnostics for all cases when a
5429 control that does not make sense is encountered.
5431 9. Added the /retain_sender option to "control=submission".
5433 10. $recipients is now available in the predata ACL (oversight).
5435 11. Tidy the search cache before the fork to do a delivery from a message
5436 received from the command line. Otherwise the child will trigger a lookup
5437 failure and thereby defer the delivery if it tries to use (for example) a
5438 cached ldap connection that the parent has called unbind on.
5440 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
5441 of $address_data from the recipient verification was clobbered by the
5442 sender verification.
5444 13. The value of address_data from a sender verification is now available in
5445 $sender_address_data in subsequent conditions in the ACL statement.
5447 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
5449 15. Added a new option "connect=<time>" to callout options, to set a different
5452 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
5453 was its contents. (It was OK if the option was not defined at all.)
5455 17. A "Completed" log line is now written for messages that are removed from
5456 the spool by the -Mrm option.
5458 18. New variables $sender_verify_failure and $recipient_verify_failure contain
5459 information about exactly what failed.
5461 19. Added -dd to debug only the daemon process.
5463 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
5464 handles timeouts, both on the server side and network timeouts. Renamed the
5465 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
5467 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
5468 It is now set to "smtps".
5470 22. $host_address is now set to the target address during the checking of
5471 ignore_target_hosts.
5473 23. When checking ignore_target_hosts for an ipliteral router, no host name was
5474 being passed; this would have caused $sender_host_name to have been used if
5475 matching the list had actually called for a host name (not very likely,
5476 since this list is usually IP addresses). A host name is now passed as
5479 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
5480 code that specifies a non-restarting handler (typically sigaction() in
5481 modern systems) in an attempt to fix a rare and obscure crash bug.
5483 25. Narrowed the window for a race in the daemon that could cause it to ignore
5484 SIGCHLD signals. This is not a major problem, because they are used only to
5485 wake it up if nothing else does.
5487 26. A malformed maildirsize file could cause Exim to calculate negative values
5488 for the mailbox size or file count. Odd effects could occur as a result.
5489 The maildirsize information is now recalculated if the size or filecount
5492 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
5493 support for a long time. Removed HAVE_SYS_VFS_H.
5495 28. Installed the latest version of exipick from John Jetmore.
5497 29. In an address list, if the pattern was not a regular expression, an empty
5498 subject address (from a bounce message) matched only if the pattern was an
5499 empty string. Non-empty patterns were not even tested. This was the wrong
5500 because it is perfectly reasonable to use an empty address as part of a
5501 database query. An empty address is now tested by patterns that are
5502 lookups. However, all the other forms of pattern expect the subject to
5503 contain a local part and a domain, and therefore, for them, an empty
5504 address still always fails if the pattern is not itself empty.
5506 30. Exim went into a mad DNS loop when attempting to do a callout where the
5507 host was specified on an smtp transport, and looking it up yielded more
5508 than one IP address.
5510 31. Re-factored the code for checking spool and log partition space into a
5511 function that finds that data and another that does the check. The former
5512 is then used to implement four new variables: $spool_space, $log_space,
5513 $spool_inodes, and $log_inodes.
5515 32. The RFC2047 encoding function was originally intended for short strings
5516 such as real names; it was not keeping to the 75-character limit for
5517 encoded words that the RFC imposes. It now respects the limit, and
5518 generates multiple encoded words if necessary. To be on the safe side, I
5519 have increased the buffer size for the ${rfc2047: expansion operator from
5522 33. It is now permitted to omit both strings after an "if" condition; if the
5523 condition is true, the result is "true". As before, when the second string
5524 is omitted, a false condition yields an empty string. This makes it less
5525 cumbersome to write custom ACL and router conditions.
5527 34. Failure to deliver a bounce message always caused it to be frozen, even if
5528 there was an errors_to setting on the router. The errors_to setting is now
5531 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
5532 canonical form (fully expanded) before being placed in
5533 $sender_host_address.
5535 36. The table in the code that translates DNS record types into text (T_A to
5536 "A" for instance) was missing entries for NS and CNAME. It is just possible
5537 that this could have caused confusion if both these types were looked up
5538 for the same domain, because the text type is used as part of Exim's
5539 per-process caching. But the chance of anyone hitting this buglet seems
5542 37. The dnsdb lookup has been extended in a number of ways.
5544 (1) There is a new type, "zns", which walks up the domain tree until it
5545 finds some nameserver records. It should be used with care.
5547 (2) There is a new type, "mxh", which is like "mx" except that it returns
5548 just the host names, not the priorities.
5550 (3) It is now possible to give a list of domains (or IP addresses) to be
5551 looked up. The behaviour when one of the lookups defers can be
5552 controlled by a keyword.
5554 (4) It is now possible to specify the separator character for use when
5555 multiple records are returned.
5557 38. The dnslists ACL condition has been extended: it is now possible to supply
5558 a list of IP addresses and/or domains to be looked up in a particular DNS
5561 39. Added log_selector=+queue_time_overall.
5563 40. When running the queue in the test harness, wait just a tad after forking a
5564 delivery process, to get repeatability of debugging output.
5566 41. Include certificate and key file names in error message when GnuTLS fails
5567 to set them up, because the GnuTLS error message doesn't include the name
5568 of the failing file when there is a problem reading it.
5570 42. Allow both -bf and -bF in the same test run.
5572 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
5574 44. The "Exiscan patch" is now merged into the mainline Exim source.
5576 45. Sometimes the final signoff response after QUIT could fail to get
5577 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
5578 < 0 before doing a fflush(). This bug looks as though it goes back to the
5579 introduction of TLS in release 3.20, but "sometimes" must have been rare
5580 because the tests only now provoked it.
5582 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
5583 (this can affect the format of dates).
5585 47. exim_tidydb, when checking for the continued existence of a message for
5586 which it has found a message-specific retry record, was not finding
5587 messages that were in split spool directories. Consequently, it was
5588 deleting retry records that should have stayed in existence.
5590 48. Steve fixed some bugs in eximstats.
5592 49. The SPA authentication driver was not abandoning authentication and moving
5593 on to the next authenticator when an expansion was forced to fail,
5594 contradicting the general specification for all authenticators. Instead it
5595 was generating a temporary error. It now behaves as specified.
5597 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
5598 (the order specifies the preference for clients). The order is now AES256,
5599 AES128, 3DES, ARCFOUR128.
5601 51. Small patch to Sieve code - explicitly set From: when generating an
5604 52. Exim crashed if a remote delivery caused a very long error message to be
5605 recorded - for instance if somebody sent an entire SpamAssassin report back
5606 as a large number of 550 error lines. This bug was coincidentally fixed by
5607 increasing the size of one of Exim's internal buffers (big_buffer) that
5608 happened as part of the Exiscan merge. However, to be on the safe side, I
5609 have made the code more robust (and fixed the comments that describe what
5612 53. Now that there can be additional text after "Completed" in log lines (if
5613 the queue_time_overall log selector is set), a one-byte patch to exigrep
5614 was needed to allow it to recognize "Completed" as not the last thing in
5617 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
5618 patch that reportedly fixes this has been added. I am not expert enough to
5619 create a test for it. This is what the patch creator wrote:
5621 "I found a little strange behaviour of ldap code when working with
5622 Windows 2003 AD Domain, where users was placed in more than one
5623 Organization Units. When I tried to give exim partial DN, the exit code
5624 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
5625 But simultaneously result of request was absolutely normal ldap result,
5626 so I produce this patch..."
5628 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
5629 REFERENCE, so I have modified the code to exclude the patch when that macro
5632 55. Some experimental protocols are using DNS PTR records for new purposes. The
5633 keys for these records are domain names, not reversed IP addresses. The
5634 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
5635 leaves it alone. Component reversal etc. now happens only for IP addresses.
5638 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
5640 57. Double the size of the debug message buffer (to 2048) so that more of very
5641 long debug lines gets shown.
5643 58. The exicyclog utility now does better if the number of log files to keep
5644 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
5646 59. Two changes related to the smtp_active_hostname option:
5648 (1) $smtp_active_hostname is now available as a variable.
5649 (2) The default for smtp_banner uses $smtp_active_hostname instead
5650 of $primary_hostname.
5652 60. The host_aton() function is supposed to be passed a string that is known
5653 to be a valid IP address. However, in the case of IPv6 addresses, it was
5654 not checking this. This is a hostage to fortune. Exim now panics and dies
5655 if the condition is not met. A case was found where this could be provoked
5656 from a dnsdb PTR lookup with an IPv6 address that had more than 8
5657 components; fortuitously, this particular loophole had already been fixed
5658 by change 4.50/55 above.
5660 If there are any other similar loopholes, the new check in host_aton()
5661 itself should stop them being exploited. The report I received stated that
5662 data on the command line could provoke the exploit when Exim was running as
5663 exim, but did not say which command line option was involved. All I could
5664 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
5665 running as the user.
5668 61. There was a buffer overflow vulnerability in the SPA authentication code
5669 (which came originally from the Samba project). I have added a test to the
5670 spa_base64_to_bits() function which I hope fixes it.
5673 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
5674 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
5676 63. The daemon start-up calls getloadavg() while still root for those OS that
5677 need the first call to be done as root, but it missed one case: when
5678 deliver_queue_load_max is set with deliver_drop_privilege. This is
5679 necessary for the benefit of the queue runner, because there is no re-exec
5680 when deliver_drop_privilege is set.
5682 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
5683 This has been fixed.
5685 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
5686 was in use, was not putting the data itself into the right store pool;
5687 consequently, it could be overwritten for a subsequent message in the same
5688 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
5691 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
5693 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
5694 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
5695 the function were treating the return as a boolean value, which happened to
5696 work because 0=false and not-0=true, but is not correct code.
5698 68. The host_aton() function was not handling scoped IPv6 addresses (those
5699 with, for example, "%eth0" on the end) correctly.
5701 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
5702 negated items (that is, ~something) in unsigned ints. Some compilers
5703 apparently mutter when there is no cast.
5705 70. If an address verification called from an ACL failed, and did not produce a
5706 user-specific message (i.e. there was only a "system" message), nothing was
5707 put in $acl_verify_message. In this situation, it now puts the system
5710 71. Change 4.23/11 added synchronization checking at the start of an SMTP
5711 session; change 4.31/43 added the unwanted input to the log line - except
5712 that it did not do this in the start of session case. It now does.
5714 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
5715 This is wrong and can cause the other end to generate a synchronization
5716 error if it is another Exim or anything else that does the synchronization
5717 check. A QUIT command is no longer sent after a timeout.
5719 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
5720 during host lookups.
5722 74. The defer_ok option of callout verification was not working if it was used
5723 when verifying addresses in header lines, that is, for this case:
5725 verify = header_sender/callout=defer_ok
5727 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
5728 those file descriptors could be used for SMTP connections. If anything
5729 wrote to stderr (the example that came up was "warn" in embedded Perl), it
5730 could be sent to the SMTP client, causing chaos. The daemon now opens
5731 stdin, stdout, and stderr to /dev/null when it puts itself into the
5734 76. Arrange for output from Perl's "warn" command to be written to Exim's main
5735 log by default. The user can override this with suitable Perl magic.
5737 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
5738 the log message when discard triggers, was not working for the DATA ACL or
5739 for the non-SMTP ACL.
5741 78. Error message wording change in sieve.c.
5743 79. If smtp_accept_max_per_host was set, the number of connections could be
5744 restricted to fewer than expected, because the daemon was trying to set up
5745 a new connection before checking whether the processes handling previous
5746 connections had finished. The check for completed processes is now done
5747 earlier. On busy systems, this bug wouldn't be noticed because something
5748 else would have woken the daemon, and it would have reaped the completed
5751 80. If a message was submitted locally by a user whose login name contained one
5752 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
5753 It caused a spool format error. I have fixed the spool reading code. A
5754 related problem was that the "from" clause in the Received: line became
5755 illegal because of the space(s). It is now covered by ${quote_local_part.
5757 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
5760 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
5762 83. Patch from Sieve maintainer to fix unterminated string problem in
5763 "vacation" handling.
5765 84. Some minor changes to the Linux configuration files to help with other
5766 OS variants using glibc.
5768 85. One more patch for Sieve to update vacation handling to latest spec.
5771 ----------------------------------------------------
5772 See the note above about the 4.44 and 4.50 releases.
5773 ----------------------------------------------------
5779 1. Change 4.43/35 introduced a bug that caused file counts to be
5780 incorrectly computed when quota_filecount was set in an appendfile
5783 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
5784 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
5787 3. Give more explanation in the error message when the command for a transport
5788 filter fails to execute.
5790 4. There are several places where Exim runs a non-Exim command in a
5791 subprocess. The SIGUSR1 signal should be disabled for these processes. This
5792 was being done only for the command run by the queryprogram router. It is
5793 now done for all such subprocesses. The other cases are: ${run, transport
5794 filters, and the commands run by the lmtp and pipe transports.
5796 5. Some older OS have a limit of 256 on the maximum number of file
5797 descriptors. Exim was using setrlimit() to set 1000 as a large value
5798 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
5799 systems. I've change it so that if it can't get 1000, it tries for 256.
5801 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
5802 was an oversight, and furthermore, ever since the addition of extra
5803 controls (e.g. 4.43/32), the checks on when to allow different forms of
5804 "control" were broken. There should now be diagnostics for all cases when a
5805 control that does not make sense is encountered.
5807 7. $recipients is now available in the predata ACL (oversight).
5809 8. Tidy the search cache before the fork to do a delivery from a message
5810 received from the command line. Otherwise the child will trigger a lookup
5811 failure and thereby defer the delivery if it tries to use (for example) a
5812 cached ldap connection that the parent has called unbind on.
5814 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
5815 of $address_data from the recipient verification was clobbered by the
5816 sender verification.
5818 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
5819 was its contents. (It was OK if the option was not defined at all.)
5821 11. A "Completed" log line is now written for messages that are removed from
5822 the spool by the -Mrm option.
5824 12. $host_address is now set to the target address during the checking of
5825 ignore_target_hosts.
5827 13. When checking ignore_target_hosts for an ipliteral router, no host name was
5828 being passed; this would have caused $sender_host_name to have been used if
5829 matching the list had actually called for a host name (not very likely,
5830 since this list is usually IP addresses). A host name is now passed as
5833 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
5834 code that specifies a non-restarting handler (typically sigaction() in
5835 modern systems) in an attempt to fix a rare and obscure crash bug.
5837 15. Narrowed the window for a race in the daemon that could cause it to ignore
5838 SIGCHLD signals. This is not a major problem, because they are used only to
5839 wake it up if nothing else does.
5841 16. A malformed maildirsize file could cause Exim to calculate negative values
5842 for the mailbox size or file count. Odd effects could occur as a result.
5843 The maildirsize information is now recalculated if the size or filecount
5846 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
5847 support for a long time. Removed HAVE_SYS_VFS_H.
5849 18. Updated exipick to current release from John Jetmore.
5851 19. Allow an empty sender to be matched against a lookup in an address list.
5852 Previously the only cases considered were a regular expression, or an
5855 20. Exim went into a mad DNS lookup loop when doing a callout where the
5856 host was specified on the transport, if the DNS lookup yielded more than
5859 21. The RFC2047 encoding function was originally intended for short strings
5860 such as real names; it was not keeping to the 75-character limit for
5861 encoded words that the RFC imposes. It now respects the limit, and
5862 generates multiple encoded words if necessary. To be on the safe side, I
5863 have increased the buffer size for the ${rfc2047: expansion operator from
5866 22. Failure to deliver a bounce message always caused it to be frozen, even if
5867 there was an errors_to setting on the router. The errors_to setting is now
5870 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
5871 canonical form (fully expanded) before being placed in
5872 $sender_host_address.
5874 24. Updated eximstats to version 1.33
5876 25. Include certificate and key file names in error message when GnuTLS fails
5877 to set them up, because the GnuTLS error message doesn't include the name
5878 of the failing file when there is a problem reading it.
5880 26. Expand error message when OpenSSL has problems setting up cert/key files.
5883 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
5884 (this can affect the format of dates).
5886 28. exim_tidydb, when checking for the continued existence of a message for
5887 which it has found a message-specific retry record, was not finding
5888 messages that were in split spool directories. Consequently, it was
5889 deleting retry records that should have stayed in existence.
5891 29. eximstats updated to version 1.35
5892 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
5893 1.35 - bugfix such that pie charts by volume are generated correctly
5895 30. The SPA authentication driver was not abandoning authentication and moving
5896 on to the next authenticator when an expansion was forced to fail,
5897 contradicting the general specification for all authenticators. Instead it
5898 was generating a temporary error. It now behaves as specified.
5900 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
5901 (the order specifies the preference for clients). The order is now AES256,
5902 AES128, 3DES, ARCFOUR128.
5904 31. Small patch to Sieve code - explicitly set From: when generating an
5907 32. Exim crashed if a remote delivery caused a very long error message to be
5908 recorded - for instance if somebody sent an entire SpamAssassin report back
5909 as a large number of 550 error lines. This bug was coincidentally fixed by
5910 increasing the size of one of Exim's internal buffers (big_buffer) that
5911 happened as part of the Exiscan merge. However, to be on the safe side, I
5912 have made the code more robust (and fixed the comments that describe what
5915 33. Some experimental protocols are using DNS PTR records for new purposes. The
5916 keys for these records are domain names, not reversed IP addresses. The
5917 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
5918 leaves it alone. Component reversal etc. now happens only for IP addresses.
5921 34. The host_aton() function is supposed to be passed a string that is known
5922 to be a valid IP address. However, in the case of IPv6 addresses, it was
5923 not checking this. This is a hostage to fortune. Exim now panics and dies
5924 if the condition is not met. A case was found where this could be provoked
5925 from a dnsdb PTR lookup with an IPv6 address that had more than 8
5926 components; fortuitously, this particular loophole had already been fixed
5927 by change 4.50/55 or 4.44/33 above.
5929 If there are any other similar loopholes, the new check in host_aton()
5930 itself should stop them being exploited. The report I received stated that
5931 data on the command line could provoke the exploit when Exim was running as
5932 exim, but did not say which command line option was involved. All I could
5933 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
5934 running as the user.
5937 35. There was a buffer overflow vulnerability in the SPA authentication code
5938 (which came originally from the Samba project). I have added a test to the
5939 spa_base64_to_bits() function which I hope fixes it.
5942 36. The daemon start-up calls getloadavg() while still root for those OS that
5943 need the first call to be done as root, but it missed one case: when
5944 deliver_queue_load_max is set with deliver_drop_privilege. This is
5945 necessary for the benefit of the queue runner, because there is no re-exec
5946 when deliver_drop_privilege is set.
5948 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
5949 was in use, was not putting the data itself into the right store pool;
5950 consequently, it could be overwritten for a subsequent message in the same
5951 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
5954 38. Sometimes the final signoff response after QUIT could fail to get
5955 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
5956 < 0 before doing a fflush(). This bug looks as though it goes back to the
5957 introduction of TLS in release 3.20, but "sometimes" must have been rare
5958 because the tests only now provoked it.
5964 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
5965 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
5966 Now it returns an integer. A number of calls were still expecting a T/F
5967 return. Fortuitously, in all cases, the tests worked in OK situations,
5968 which is the norm. However, things would have gone wrong on any write
5969 failures on the smtp file descriptor. This function is used when sending
5970 messages over SMTP and also when doing verify callouts.
5972 2. When Exim is called to do synchronous delivery of a locally submitted
5973 message (the -odf or -odi options), it no longer closes stderr before doing
5976 3. Implemented the mua_wrapper option.
5978 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
5980 5. Implemented the functions header_remove(), header_testname(),
5981 header_add_at_position(), and receive_remove_recipient(), and exported them
5984 6. If an ACL "warn" statement specified the addition of headers, Exim already
5985 inserted X-ACL-Warn: at the start if there was no header name. However, it
5986 was not making this test for the second and subsequent header lines if
5987 there were newlines in the string. This meant that an invalid header could
5988 be inserted if Exim was badly configured.
5990 7. Allow an ACL "warn" statement to add header lines at the start or after all
5991 the Received: headers, as well as at the end.
5993 8. Added the rcpt_4xx retry error code.
5995 9. Added postmaster_mailfrom=xxx to callout verification option.
5997 10. Added mailfrom=xxxx to the callout verification option, for verify=
6000 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
6001 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
6003 12. Inserted some casts to stop certain compilers warning when using pointer
6004 differences as field lengths or precisions in printf-type calls (mostly
6005 affecting debugging statements).
6007 13. Added optional readline() support for -be (dynamically loaded).
6009 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
6010 same clock tick as a message's arrival, so that its received time was the
6011 same as the "first fail" time on the retry record, and that message
6012 remained on the queue past the ultimate address timeout, every queue runner
6013 would try a delivery (because it was past the ultimate address timeout) but
6014 after another failure, the ultimate address timeout, which should have then
6015 bounced the address, did not kick in. This was a "< instead of <=" error;
6016 in most cases the first failure would have been in the next clock tick
6017 after the received time, and all would be well.
6019 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
6020 being recognized when the domain list was tested by the match_domain
6021 condition in an expansion string.
6023 16. Added the ${str2b64: operator.
6025 17. Exim was always calling setrlimit() to set a large limit for the number of
6026 processes, without checking whether the existing limit was already
6027 adequate. (It did check for the limit on file descriptors.) Furthermore,
6028 errors from getrlimit() and setrlimit() were being ignored. Now they are
6029 logged to the main and panic logs, but Exim does carry on, to try to do its
6030 job under whatever limits there are.
6032 18. Imported PCRE 5.0.
6034 19. Trivial typo in log message " temporarily refused connection" (the leading
6037 20. If the log selector return_path_on_delivery was set and an address was
6038 redirected to /dev/null, the delivery process crashed because it assumed
6039 that a return path would always be set for a "successful" delivery. In this
6040 case, the whole delivery is bypassed as an optimization, and therefore no
6043 21. Internal re-arrangement: the function for sending a challenge and reading
6044 a response while authentication was assuming a zero-terminated challenge
6045 string. It's now changed to take a pointer and a length, to allow for
6046 binary data in such strings.
6048 22. Added the cyrus_sasl authenticator (code supplied by MBM).
6050 23. Exim was not respecting finduser_retries when seeking the login of the
6051 uid under which it was called; it was always trying 10 times. (The default
6052 setting of finduser_retries is zero.) Also, it was sleeping after the final
6053 failure, which is pointless.
6055 24. Implemented tls_on_connect_ports.
6057 25. Implemented acl_smtp_predata.
6059 26. If the domain in control=submission is set empty, Exim assumes that the
6060 authenticated id is a complete email address when it generates From: or
6061 Sender: header lines.
6063 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
6064 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
6065 chown and chgrp in /bin and hostname in /usr/bin.
6067 28. Exim was keeping the "process log" file open after each use, just as it
6068 does for the main log. This opens the possibility of it remaining open for
6069 long periods when the USR1 signal hits a daemon. Occasional processlog
6070 errors were reported, that could have been caused by this. Anyway, it seems
6071 much more sensible not to leave this file open at all, so that is what now
6074 29. The long-running daemon process does not normally write to the log once it
6075 has entered its main loop, and it closes the log before doing so. This is
6076 so that log files can straightforwardly be renamed and moved. However,
6077 there are a couple of unusual error situations where the daemon does write
6078 log entries, and I had neglected to close the log afterwards.
6080 30. The text of an SMTP error response that was received during a remote
6081 delivery was being truncated at 512 bytes. This is too short for some of
6082 the long messages that one sometimes sees. I've increased the limit to
6085 31. It is now possible to make retry rules that apply only when a message has a
6086 specific sender, in particular, an empty sender.
6088 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
6089 it possible to be selective about when SMTP synchronization is enforced.
6091 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
6093 32. Implemented hosts_connection_nolog.
6095 33. Added an ACL for QUIT.
6097 34. Setting "delay_warning=" to disable warnings was not working; it gave a
6100 35. Added mailbox_size and mailbox_filecount to appendfile.
6102 36. Added control = no_multiline_responses to ACLs.
6104 37. There was a bug in the logic of the code that waits for the clock to tick
6105 in the case where the clock went backwards by a substantial amount such
6106 that the microsecond fraction of "now" was more than the microsecond
6107 fraction of "then" (but the whole seconds number was less).
6109 38. Added support for the libradius Radius client library this is found on
6110 FreeBSD (previously only the radiusclient library was supported).
6116 1. When certain lookups returned multiple values in the form name=value, the
6117 quoting of the values was not always being done properly. Specifically:
6118 (a) If the value started with a double quote, but contained no whitespace,
6120 (b) If the value contained whitespace other than a space character (i.e.
6121 tabs or newlines or carriage returns) it was not quoted.
6122 This fix has been applied to the mysql and pgsql lookups by writing a
6123 separate quoting function and calling it from the lookup code. The fix
6124 should probably also be applied to nisplus, ibase and oracle lookups, but
6125 since I cannot test any of those, I have not disturbed their existing code.
6127 2. A hit in the callout cache for a specific address caused a log line with no
6128 reason for rejecting RCPT. Now it says "Previous (cached) callout
6129 verification failure".
6131 3. There was an off-by-one bug in the queryprogram router. An over-long
6132 return line was truncated at 256 instead of 255 characters, thereby
6133 overflowing its buffer with the terminating zero. As well as fixing this, I
6134 have increased the buffer size to 1024 (and made a note to document this).
6136 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
6137 when Exim is waiting for an SMTP response from a remote server, Exim
6138 restarts its select() call on the socket, thereby resetting its timeout.
6139 This is not a problem when such interrupts are rare. Somebody set up a cron
6140 job to run exiwhat every 2 minutes, which is less than the normal select()
6141 timeout (5 or 10 minutes). This meant that the select() timeout never
6142 kicked in because it was always reset. I have fixed this by comparing the
6143 time when an interrupt arrives with the time at the start of the first call
6144 to select(). If more time than the timeout has elapsed, the interrupt is
6145 treated as a timeout.
6147 5. Some internal re-factoring in preparation for the addition of Sieve
6148 extensions (by MH). In particular, the "personal" test is moved to a
6149 separate function, and given an option for scanning Cc: and Bcc: (which is
6150 not set for Exim filters).
6152 6. When Exim created an email address using the login of the caller as the
6153 local part (e.g. when creating a From: or Sender: header line), it was not
6154 quoting the local part when it contained special characters such as @.
6156 7. Installed new OpenBSD configuration files.
6158 8. Reworded some messages for syntax errors in "and" and "or" conditions to
6159 try to make them clearer.
6161 9. Callout options, other than the timeout value, were being ignored when
6162 verifying sender addresses in header lines. For example, when using
6164 verify = header_sender/callout=no_cache
6166 the cache was (incorrectly) being used.
6168 10. Added a missing instance of ${EXE} to the exim_install script; this affects
6169 only the Cygwin environment.
6171 11. When return_path_on_delivery was set as a log selector, if different remote
6172 addresses in the same message used different return paths and parallel
6173 remote delivery occurred, the wrong values would sometimes be logged.
6174 (Whenever a remote delivery process finished, the return path value from
6175 the most recently started remote delivery process was logged.)
6177 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
6178 lines when AUTH and/or TLS are in use. This is the "received protocol"
6179 field. Exim used to use "asmtp" for authenticated SMTP, without any
6180 indication (in the protocol name) for TLS use. Now it follows the RFC and
6181 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
6182 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
6183 names appear in log lines as well as in Received: header lines.
6185 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
6186 extensions, and comparison tests, and to fix some bugs.
6188 14. Changes to the "personal" filter test:
6190 (1) The test was buggy in that it was just doing the equivalent of
6191 "contains" tests on header lines. For example, if a user's address was
6192 anne@some.where, the "personal" test would incorrectly be true for
6194 To: susanne@some.where
6196 This test is now done by extracting each address from the header in turn,
6197 and checking the entire address. Other tests that are part of "personal"
6198 are now done using regular expressions (for example, to check local parts
6199 of addresses in From: header lines).
6201 (2) The list of non-personal local parts in From: addresses has been
6202 extended to include "listserv", "majordomo", "*-request", and "owner-*",
6203 taken from the Sieve specification recommendations.
6205 (3) If the message contains any header line starting with "List-" it is
6206 treated as non-personal.
6208 (4) The test for "circular" in the Subject: header line has been removed
6209 because it now seems ill-conceived.
6211 15. Minor typos in src/EDITME comments corrected.
6213 16. Installed latest exipick from John Jetmore.
6215 17. If headers_add on a router specified a text string that was too long for
6216 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
6217 of string_sprintf() is now avoided.
6219 18. $message_body_size was not set (it was always zero) when running the DATA
6220 ACL and the local_scan() function.
6222 19. For the "mail" command in an Exim filter, no default was being set for
6223 the once_repeat time, causing a random time value to be used if "once" was
6224 specified. (If the value happened to be <= 0, no repeat happened.) The
6225 default is now 0s, meaning "never repeat". The "vacation" command was OK
6226 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
6227 (I found it when inspecting the code).
6229 20. There is now an overall timeout for performing a callout verification. It
6230 defaults to 4 times the callout timeout, which applies to individual SMTP
6231 commands during the callout. The overall timeout applies when there is more
6232 than one host that can be tried. The timeout is checked before trying the
6233 next host. This prevents very long delays if there are a large number of
6234 hosts and all are timing out (e.g. when the network connections are timing
6235 out). The value of the overall timeout can be changed by specifying an
6236 additional sub-option for "callout", called "maxwait". For example:
6238 verify = sender/callout=5s,maxwait=20s
6240 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
6241 to the end before writing, but this should make it even safer).
6243 22. Exim was forgetting that it had advertised PIPELINING for the second and
6244 subsequent messages on an SMTP connection. It was also not resetting its
6245 memory on STARTTLS and an internal HELO.
6247 23. When Exim logs an SMTP synchronization error within a session, it now
6248 records whether PIPELINING has been advertised or not.
6250 24. Added 3 instances of "(long int)" casts to time_t variables that were being
6251 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
6252 rather than long int.
6254 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
6256 26. Added the never_mail option to autoreply.
6262 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
6263 crash if the getsockname() call failed; this can happen if a connection is
6264 closed very soon after it is established. The problem was simply in the
6265 order in which certain operations were done, causing Exim to try to write
6266 to the SMTP stream before it had set up the file descriptor. The bug has
6267 been fixed by making things happen in the correct order.
6273 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
6274 before the connection was closed, thus losing the rejection response.
6276 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
6277 some early Solaris releases, but causes trouble in current releases where
6278 socklen_t is defined.
6280 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
6283 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
6286 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
6287 that releasing the top of it at the end releases what was used for sub-
6288 expansions (unless the block got too big). However, discard this block if
6289 the first thing is a variable or header, so that we can use its block when
6290 it is dynamic (useful for very large $message_headers, for example).
6292 6. Lookups now cache *every* query, not just the most recent. A new, separate
6293 store pool is used for this. It can be recovered when all lookup caches are
6294 flushed. Lookups now release memory at the end of their result strings.
6295 This has involved some general refactoring of the lookup sources.
6297 7. Some code has been added to the store_xxx() functions to reduce the amount
6298 of flapping under certain conditions.
6300 8. log_incoming_interface used to affect only the <= reception log lines. Now
6301 it causes the local interface and port to be added to several more SMTP log
6302 lines, for example "SMTP connection from", and rejection lines.
6304 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
6306 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
6308 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
6309 could be overwritten at the end of the current message (or the start of a
6310 new message if it was set in a HELO ACL). The value is now preserved for
6311 the duration of the SMTP connection.
6313 12. If a transport had a headers_rewrite setting, and a matching header line
6314 contained an unqualified address, that address was qualified, even if it
6315 did not match any rewriting rules. The underlying bug was that the values
6316 of the flags that permit the existence of unqualified sender and recipient
6317 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
6318 non-local messages, and by -bnq for local messages) were not being
6319 preserved with the message after it was received.
6321 13. When Exim was logging an SMTP synchronization error, it could sometimes log
6322 "next input=" as part of the text comprising the host identity instead of
6323 the correct text. The code was using the same buffer for two different
6324 strings. However, depending on which order the printing function evaluated
6325 its arguments, the bug did not always show up. Under Linux, for example, my
6326 test suite worked just fine.
6328 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
6329 doesn't work with some older versions of Perl. It has been changed to "my",
6330 which in any case is probably the better facility to use.
6332 15. A really picky compiler found some instances of statements for creating
6333 error messages that either had too many or two few arguments for the format
6336 16. The size of the buffer for calls to the DNS resolver has been increased
6337 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
6338 for addresses that have a lot of PTR records. This alleviates a problem; it
6339 does not fully solve it.
6341 17. A dnsdb lookup for PTR records that receives more data than will fit in the
6342 buffer now truncates the list and logs the incident, which is the same
6343 action as happens when Exim is looking up a host name and its aliases.
6344 Previously in this situation something unpredictable would happen;
6345 sometimes it was "internal error: store_reset failed".
6347 18. If a server dropped the connection unexpectedly when an Exim client was
6348 using GnuTLS and trying to read a response, the client delivery process
6349 crashed while trying to generate an error log message.
6351 19. If a "warn" verb in an ACL added multiple headers to a message in a single
6352 string, for example:
6354 warn message = H1: something\nH2: something
6356 the text was added as a single header line from Exim's point of view
6357 though it ended up OK in the delivered message. However, searching for the
6358 second and subsequent header lines using $h_h2: did not work. This has been
6359 fixed. Similarly, if a system filter added multiple headers in this way,
6360 the routers could not see them.
6362 20. Expanded the error message when iplsearch is called with an invalid key to
6363 suggest using net-iplsearch in a host list.
6365 21. When running tests using -bh, any delays imposed by "delay" modifiers in
6366 ACLs are no longer actually imposed (and a message to that effect is
6369 22. If a "gecos" field in a passwd entry contained escaped characters, in
6370 particular, if it contained a \" sequence, Exim got it wrong when building
6371 a From: or a Sender: header from that name. A second bug also caused
6372 incorrect handling when an unquoted " was present following a character
6373 that needed quoting.
6375 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
6376 was not being matched caselessly.
6378 24. Arranged for all hyphens in the exim.8 source to be escaped with
6381 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
6382 properly. Recipient callout cache records were still being keyed to include
6383 the sender, even when use_sender was set false. This led to far more
6384 callouts that were necessary. The sender is no longer included in the key
6385 when use_sender is false.
6387 26. Added "control = submission" modifier to ACLs.
6389 27. Added the ${base62d: operator to decode base 62 numbers.
6391 28. dnsdb lookups can now access SRV records.
6393 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
6394 the configuration file.
6396 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
6397 (-v) mode. This makes the output for a verbose queue run more intelligible.
6399 31. Added a use_postmaster feature to recipient callouts.
6401 32. Added the $body_zerocount variable, containing the number of binary zero
6402 bytes in the message body.
6404 33. The time of last modification of the "new" subdirectory is now used as the
6405 "mailbox time last read" when there is a quota error for a maildir
6408 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
6410 35. Added +ignore_unknown as a special item in host lists.
6412 36. Code for decoding IPv6 addresses in host lists is now included, even if
6413 IPv6 support is not being compiled. This fixes a bug in which an IPv6
6414 address was recognized as an IP address, but was then not correctly decoded
6415 into binary, causing unexpected and incorrect effects when compared with
6422 1. Very minor rewording of debugging text in manualroute to say "list of
6423 hosts" instead of "hostlist".
6425 2. If verify=header_syntax was set, and a header line with an unqualified
6426 address (no domain) and a large number of spaces between the end of the
6427 name and the colon was received, the reception process suffered a buffer
6428 overflow, and (when I tested it) crashed. This was caused by some obsolete
6429 code that should have been removed. The fix is to remove it!
6431 3. When running in the test harness, delay a bit after writing a bounce
6432 message to get a bit more predictability in the log output.
6434 4. Added a call to search_tidyup() just before forking a reception process. In
6435 theory, someone could use a lookup in the expansion of smtp_accept_max_
6436 per_host which, without the tidyup, could leave open a database connection.
6438 5. Added the variables $recipient_data and $sender_data which get set from a
6439 lookup success in an ACL "recipients" or "senders" condition, or a router
6440 "senders" option, similar to $domain_data and $local_part_data.
6442 6. Moved the writing of debug_print from before to after the "senders" test
6445 7. Change 4.31/66 (moving the time when the Received: is generated) caused
6446 problems for message scanning, either using a data ACL, or using
6447 local_scan() because the Received: header was not generated till after they
6448 were called (in order to set the time as the time of reception completion).
6449 I have revised the way this works. The header is now generated after the
6450 body is received, but before the ACL or local_scan() are called. After they
6451 are run, the timestamp in the header is updated.
6457 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
6458 before starting a queue runner without re-exec. This happened only when
6459 deliver_drop_privilege was set or when the Exim user was set to root. The
6460 effect of the bug was that timeouts during subsequent deliveries caused
6461 crashes instead of being properly handled. The handler is now left at its
6462 default (and expected) setting.
6464 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
6465 message, again when deliver_drop_privilege is set or Exim is run as root.
6466 The bug described in (1) was not present in this case, but the tidying up
6467 of the other signals was missing. I have made the two cases consistent.
6469 3. The ignore_target_hosts setting on a manualroute router was being ignored
6470 for hosts that were looked up using the /MX notation.
6472 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
6475 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
6476 operated on the sender address. After changing the $sender_address to <>
6477 for the sender address verify, Exim was re-instated it as the original
6478 (before rewriting) address, but remembering that it had rewritten it, so it
6479 wasn't rewriting it again. This bug also had the effect of breaking the
6480 sender address verification caching when the sender address was rewritten.
6482 6. The ignore_target_hosts option was being ignored by the ipliteral router.
6483 This has been changed so that if the ip literal address matches
6484 ignore_target_hosts, the router declines.
6486 7. Added expansion conditions match_domain, match_address, and match_local_
6487 part (NOT match_host).
6489 8. The placeholder for the Received: header didn't have a length field set.
6491 9. Added code to Exim itself and to exim_lock to test for a specific race
6492 condition that could lead to file corruption when using MBX delivery. The
6493 issue is with the lockfile that is created in /tmp. If this file is removed
6494 after a process has opened it but before that process has acquired a lock,
6495 there is the potential for a second process to recreate the file and also
6496 acquire a lock. This could lead to two Exim processes writing to the file
6497 at the same time. The added code performs the same test as UW imapd; it
6498 checks after acquiring the lock that its file descriptor still refers to
6499 the same named file.
6501 10. The buffer for building added header lines was of fixed size, 8192 bytes.
6502 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
6505 11. Added the smtp_active_hostname option. If used, this will typically be made
6506 to depend on the incoming interface address. Because $interface_address is
6507 not set up until the daemon has forked a reception process, error responses
6508 that can happen earlier (such as "too many connections") no longer contain
6511 12. If an expansion in a condition on a "warn" statement fails because a lookup
6512 defers, the "warn" statement is abandoned, and the next ACL statement is
6513 processed. Previously this caused the whole ACL to be aborted.
6515 13. Added the iplsearch lookup type.
6517 14. Added ident_timeout as a log selector.
6519 15. Added tls_certificate_verified as a log selector.
6521 16. Added a global option tls_require_ciphers (compare the smtp transport
6522 option of the same name). This controls incoming TLS connections.
6524 17. I finally figured out how to make tls_require_ciphers do a similar thing
6525 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
6526 before starting the TLS session.
6528 18. Tabs are now shown as \t in -bP output.
6530 19. If the log selector return_path_on_delivery was set, Exim crashed when
6531 bouncing a message because it had too many Received: header lines.
6533 20. If two routers both had headers_remove settings, and the first one included
6534 a superfluous trailing colon, the final name in the first list and the
6535 first name in the second list were incorrectly joined into one item (with a
6536 colon in the middle).
6542 1. Added -C and -D options to the exinext utility, mainly to make it easier
6543 to include in the automated testing, but these could be helpful when
6544 multiple configurations are in use.
6546 2. The exinext utility was not formatting the output nicely when there was
6547 an alternate port involved in the retry record key, nor when there was a
6548 message id as well (for retries that were specific to a specific message
6549 and a specific host). It was also confused by IPv6 addresses, because of
6550 the additional colons they contain. I have fixed the IPv4 problem, and
6551 patched it up to do a reasonable job for IPv6.
6553 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
6554 delivery, the log line now contains "pipelined" if PIPELINING was used.
6556 4. An SMTP transport process used to panic and die if the bind() call to set
6557 an explicit outgoing interface failed. This has been changed; it is now
6558 treated in the same way as a connect() failure.
6560 5. A reference to $sender_host_name in the part of a conditional expansion
6561 that was being skipped was still causing a DNS lookup. This no longer
6564 6. The def: expansion condition was not recognizing references to header lines
6565 that used bh_ and bheader_.
6567 7. Added the _cache feature to named lists.
6569 8. The code for checking quota_filecount in the appendfile transport was
6570 allowing one more file than it should have been.
6572 9. For compatibility with Sendmail, the command line option
6580 and sets the incoming protocol and host name (for trusted callers). The
6581 host name and its colon can be omitted when only the protocol is to be set.
6582 Note the Exim already has two private options, -pd and -ps, that refer to
6583 embedded Perl. It is therefore impossible to set a protocol value of "d" or
6584 "s", but I don't think that's a major issue.
6586 10. A number of refactoring changes to the code, none of which should affect
6589 (a) The number of logging options was getting close to filling up the
6590 32-bit word that was used as a bit map. I have split them into two classes:
6591 those that are passed in the argument to log_write(), and those that are
6592 only ever tested independently outside of that function. These are now in
6593 separate 32-bit words, so there is plenty of room for expansion again.
6594 There is no change in the user interface or the logging behaviour.
6596 (b) When building, for example, log lines, the code previously used a
6597 macro that called string_cat() twice, in order to add two strings. This is
6598 not really sufficiently general. Furthermore, there was one instance where
6599 it was actually wrong because one of the argument was used twice, and in
6600 one call a function was used. (As it happened, calling the function twice
6601 did not affect the overall behaviour.) The macro has been replaced by a
6602 function that can join an arbitrary number of extra strings onto a growing
6605 (c) The code for expansion conditions now uses a table and a binary chop
6606 instead of a serial search (which was left over from when there were very
6607 few conditions). Also, it now recognizes conditions like "pam" even when
6608 the relevant support is not compiled in: a suitably worded error message is
6609 given if an attempt is made to use such a condition.
6611 11. Added ${time_interval:xxxxx}.
6613 12. A bug was causing one of the ddress fields not to be passed back correctly
6614 from remote delivery subprocesses. The field in question was not being
6615 subsequently used, so this caused to problems in practice.
6617 13. Added new log selectors queue_time and deliver_time.
6619 14. Might have fixed a bug in maildirsizefile handling that threw up
6620 "unexpected character" debug warnings, and recalculated the data
6621 unnecessarily. In any case, I expanded the warning message to give more
6624 15. Added the message "Restricted characters in address" to the statements in
6625 the default ACL that block characters like @ and % in local parts.
6627 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
6628 Three changes have been made:
6630 (a) There was a serious bug; a negative response to MAIL caused the whole
6631 recipient domain to be cached as invalid, thereby blocking all messages
6632 to all local parts at the same domain, from all senders. This bug has
6633 been fixed. The domain is no longer cached after a negative response to
6634 MAIL if the sender used is not empty.
6636 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
6639 (c) A new callout option, "use_sender" has been added for people who want
6640 the modified behaviour.
6646 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
6649 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
6650 indeed breaks things for older releases.
6652 3. Added additional logging to the case where there is a problem reading data
6653 from a filter that is running in a subprocess using a pipe, in order to
6654 try to track down a specific problem.
6656 4. Testing facility fudge: when running in the test harness and attempting
6657 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
6658 getting "No route to host". Convert this to a timeout.
6660 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
6663 6. Some OS don't have socklen_t but use size_t instead. This affects the
6664 fifth argument of getsockopt() amongst other things. This is now
6665 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
6666 can be set for individual OS. I have set it for SunOS5, OSF1, and
6667 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
6668 some earlier ones do not.
6670 7. Change 4.30/15 was not doing the test caselessly.
6672 8. The standard form for an IPv6 address literal was being rejected by address
6673 parsing in, for example, MAIL and RCPT commands. An example of this kind of
6674 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6675 this, as well as the form without the "IPv6" on the front (but only when
6676 address literals are enabled, of course).
6678 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
6680 10. Exim crashed if a message with an empty sender address specified by -f
6681 encountered a router with an errors_to setting. This could be provoked only
6682 by a command such as
6686 where an empty string was supplied; "<>" did not hit this bug.
6688 11. Installed PCRE release 4.5.
6690 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
6691 remained set. It is now erased.
6693 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
6694 times from message ids (which are base 36 rather than the normal 62).
6696 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
6697 were being counted as actual protocol errors, and logged if the log
6698 selector +smtp_protocol_error was set. One cannot be perfect in this test,
6699 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
6700 and DATA following a set of rejected RCPTs do not count as protocol errors.
6701 In other words, Exim assumes they were pipelined, though this may not
6702 actually be the case. Of course, in all cases the client gets an
6703 appropriate error code.
6705 15. If a lookup fails in an ACL condition, a message about the failure may
6706 be available; it is used if testing the ACL cannot continue, because most
6707 such messages specify what the cause of the deferral is. However, some
6708 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
6709 that caused an old message to be retained and used if a later statement
6710 caused a defer, replacing the real cause of the deferral.
6712 16. If an IP address had so many PTR records that the DNS lookup buffer
6713 was not large enough to hold them, Exim could crash while trying to process
6714 the truncated data. It now detects and logs this case.
6716 17. Further to 4.21/58, another change has been made: if (and only if) the
6717 first line of a message (the first header line) ends with CRLF, a bare LF
6718 in a subsequent header line has a space inserted after it, so as not to
6719 terminate the header.
6721 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
6722 unnecessarily, used atoi() instead of strtol(), and didn't check the
6723 termination when getting file sizes from file names by regex.
6725 19. Completely re-implemented the support for maildirsize files, in the light
6726 of a number of problems with the previous contributed implementation
6727 (4.30/29). In particular:
6729 . If the quota is zero, the maildirsize file is maintained, but no quota is
6732 . If the maildir directory does not exist, it is created before any attempt
6733 to write a maildirsize file.
6735 . The quota value in the file is just a cache; if the quota is changed in
6736 the transport, the new value overrides.
6738 . A regular expression is available for excluding directories from the
6741 20. The autoreply transport checks the characters in options that define the
6742 message's headers; it allows continued headers, but it was checking with
6743 isspace() after an embedded newline instead of explicitly looking for a
6746 21. If all the "regular" hosts to which an address was routed had passed their
6747 expiry times, and had not reached their retry times, the address was
6748 bounced, even if fallback hosts were defined. Now Exim should go on to try
6751 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
6752 equivalent code in the SMTP transport. Some hosts send humungous responses
6753 to HELO/EHLO, more than 1024 it seems.
6755 23. Refactoring: code in filter.c used (void *) for "any old type" but this
6756 gives compiler warnings in some environments. I've now done it "properly",
6759 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
6760 (because of problems with the built-in one) was declared to return uschar *
6761 instead of char *, causing compiler failure.
6763 25. Fixed a file descriptor leak when processing alias/forward files.
6765 26. Fixed a minor format string issue in dbfn.c.
6767 27. Typo in exim.c: ("dmbnz" for "dbmnz").
6769 28. If a filter file refered to $h_xxx or $message_headers, and the headers
6770 contained RFC 2047 "words", Exim's memory could, under certain conditions,
6773 29. When a sender address is verified, it is cached, to save repeating the test
6774 when there is more than one recipient in a message. However, when the
6775 verification involves a callout, it is possible for different callout
6776 options to be set for different recipients. It is too complicated to keep
6777 track of this in the cache, so now Exim always runs a verification when a
6778 callout is required, relying on the callout cache for the optimization.
6779 The overhead is duplication of the address routing, but this should not be
6782 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
6783 to be verified with callout=postmaster, and the main callout worked but the
6784 postmaster check failed, the verification correctly failed. However, if a
6785 subsequent RCPT command asked for sender verification *without* the
6786 postmaster check, incorrect caching caused this verification also to fail,
6789 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
6790 it was not caching the DNS options (qualify_single, search_parents) that
6791 were used when the lookup failed. A subsequent lookup with different
6792 options therefore always gave the same answer, though there were cases
6793 where it should not have. (Example: a "domains = !$mx_any" option on a
6794 dnslookup router: the "domains" option is always processed without any
6795 widening, but the router might have qualify_single set.) Now Exim uses the
6796 cached value only when the same options are set.
6798 32. Added John Jetmore's "exipick" utility to the distribution.
6800 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
6801 than a timeout (e.g. a certificate is required, and is not provided), an
6802 Exim server now closes the connection immediately. Previously it waited for
6803 the client to close - but if the client is SSL, it seems that they each
6804 wait for each other, leading to a delay before one of them times out.
6806 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
6807 maintained 0.8.x compatibility because I don't think many are using it, and
6808 it is clearly obsolete.
6810 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
6813 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
6814 client certificate was expired. A simple patch fixes this, though I don't
6815 understand the full logic of why the verify callback is called multiple
6818 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
6819 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
6820 which causes problems with some clients (such as the Certicom SSL Plus
6821 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
6822 disables the coutermeasure allowing Eudora to connect."
6824 38. Exim was not checking that a write() to a log file succeeded. This could
6825 lead to Bad Things if a log got too big, in particular if it hit a file
6826 size limit. Exim now panics and dies if it cannot write to a log file, just
6827 as it does if it cannot open a log file.
6829 39. Modified OS/Makefile-Linux so that it now contains
6831 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
6833 The two -D definitions ensure that Exim is compiled with large file
6834 support, which makes it possible to handle log files that are bigger than
6837 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
6838 instance) a domain was checked against a named list that involved a lookup,
6839 causing $domain_data to be set, then another domain was checked against the
6840 same list, then the first domain was re-checked, the value of $domain_data
6841 after the final check could be wrong. In particular, if the second check
6842 failed, it could be set empty. This bug probably also applied to
6845 41. The strip_trailing_dot option was not being applied to the address given
6846 with the -f command-line option.
6848 42. The code for reading a message's header from the spool was incrementing
6849 $received_count, but never initializing it. This meant that the value was
6850 incorrect (doubled) while delivering a message in the same process in which
6851 it was received. In the most common configuration of Exim, this never
6852 happens - a fresh exec is done - but it can happen when
6853 deliver_drop_privilege is set.
6855 43. When Exim logs an SMTP synchronization error - client data sent too soon -
6856 it now includes up to 150 characters of the unexpected data in the log
6859 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
6860 and building data strings. The size of both of these buffers was 10 000
6861 bytes - far larger than anybody would *ever* want, thought I. Needless to
6862 say, somebody hit the limit. I have increased the maximum line length to
6863 20 000 and the maximum data length of concatenated lines to 100 000. I have
6864 also fixed two bugs, because there was no checking on these buffers. Tsk,
6865 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
6866 buffer is too small.
6868 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
6869 lsearch lookups. Now it does.
6871 46. When parsing a route_list item in a manualroute router, a fixed-length
6872 buffer was used for the list of hosts. I made this 1024 bytes long,
6873 thinking that nobody would ever have a list of hosts that long. Wrong.
6874 Somebody had a whole pile of complicated expansion conditions, and the
6875 string was silently truncated, leading to an expansion error. It turns out
6876 that it is easier to change to an unlimited length (owing to other changes
6877 that have happened since this code was originally written) than to build
6878 structure for giving a limitation error. The length of the item that
6879 expands into the list of hosts is now unlimited.
6881 47. The lsearch lookup could not handle data where the length of text line was
6882 more than 4095 characters. Such lines were truncated, leading to shortened
6883 data being returned. It should now handle lines of any length.
6885 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
6886 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
6889 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
6890 replacing strings such as BIN_DIRECTORY in a source file: the replacement
6891 no longer happens in comment lines. A list of replacements is now placed
6892 at the head of all of the source files, except those whose only change is
6893 to replace PERL_COMMAND in the very first #! line.
6895 50. Replaced the slow insertion sort in queue.c, for sorting the list of
6896 messages on the queue, with a bottom-up merge sort, using code contributed
6897 by Michael Haardt. This should make operations like -bp somewhat faster on
6898 large queues. It won't affect queue runners, except when queue_run_in_order
6901 51. Installed eximstats 1.31 in the distribution.
6903 52. Added support for SRV lookups to the dnslookup router.
6905 53. If an ACL referred to $message_body or $message_body_end, the value was not
6906 reset for any messages that followed in the same SMTP session.
6908 54. The store-handling optimization for building very long strings was not
6909 differentiating between the different store pools. I don't think this
6910 actually made any difference in practice, but I've tidied it.
6912 55. While running the routers to verify a sender address, $sender_address
6913 was still set to the sender address. This is wrong, because when routing to
6914 send a bounce to the sender, it would be empty. Therefore, I have changed
6915 it so that, while verifying a sender address, $sender_address is set to <>.
6916 (There is no change to what happens when verifying a recipient address.)
6918 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
6919 target A or AAAA records (if not already returned) without resetting the
6920 qualify_single or search_parents options of the DNS resolver. These are
6921 inappropriate in this case because the targets of MX and SRV records must
6922 be FQDNs. A broken DNS record could cause trouble if it happened to have a
6923 target that, when qualified, matched something in the local domain. These
6924 two options are now turned off when doing these lookups.
6926 57. It seems that at least some releases of Reiserfs (which does not have the
6927 concept of a fixed number of inodes) returns zero and not -1 for the
6928 number of available inodes. This interacted badly with check_spool_inodes,
6929 which assumed that -1 was the "no such thing" setting. What I have done is
6930 to check that the total number of inodes is greater than zero before doing
6931 the test of how many are available.
6933 58. When a "warn" ACL statement has a log_message modifier, the message is
6934 remembered, and not repeated. This is to avoid a lot of repetition when a
6935 message has many recipients that cause the same warning to be written.
6936 However, Exim was preserving the list of already written lines for an
6937 entire SMTP session, which doesn't seem right. The memory is now reset if a
6938 new message is started.
6940 59. The "rewrite" debugging flag was not showing the result of rewriting in the
6941 debugging output unless log_rewrite was also set.
6943 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
6944 of (int)(handle) when we know that handle contains (void *)(-1).
6946 61. The Exim daemon panic-logs an error return when it closes the incoming
6947 connection. However "connection reset by peer" seems to be common, and
6948 isn't really an error worthy of noting specially, so that particular error
6951 62. When Exim is trying to find all the local interfaces, it used to panic and
6952 die if the ioctl to get the interface flags failed. However, it seems that
6953 on at least one OS (Solaris 9) it is possible to have an interface that is
6954 included in the list of interfaces, but for which you get a failure error
6955 for this call. This happens when the interface is not "plumbed" into a
6956 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
6957 failure of the "get flags" call assumes that the interface is down.
6959 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
6960 makes life easier for people who are doing arithmetic on fields extracted
6961 from dates, where you often get leading zeros that should not be
6962 interpreted as octal.
6964 64. Added qualify_domain to the redirect router, to override the global
6967 65. If a pathologically long header line contained very many addresses (the
6968 report of this problem mentioned 10 000) and each of them was rewritten,
6969 Exim could use up a very large amount of memory. (It kept on making new
6970 copies of the header line as it rewrote, and never released the old ones.)
6971 At the expense of a bit more processing, the header rewriting function has
6972 been changed so that it no longer eats memory in this way.
6974 66. The generation of the Received: header has been moved from the time that a
6975 message starts to be received, to the time that it finishes. The timestamp
6976 in the Received: header should now be very close to that of the <= log
6977 line. There are two side-effects of this change:
6979 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
6980 logged header lines no longer include the local Received: line, because
6981 it has not yet been created. The same applies to a copy of the message
6982 that is returned to a non-SMTP sender when a message is rejected.
6984 (b) When a filter file is tested using -bf, no additional Received: header
6985 is added to the test message. After some thought, I decided that this
6988 This change does not affect the value of $received_for. It is still set
6989 after address rewriting, but before local_scan() is called.
6991 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
6993 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
6994 gave an unhelpful panic error message, and a defer error. I have managed to
6995 change this behaviour so that it now rejects any supplied certificate,
6996 which seems right, as the list of acceptable certificates is empty.
6998 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
6999 gave an unhelpful defer error. I have not managed to make this reject any
7000 supplied certificates, but the error message it gives is "no certificate
7001 supplied", which is not helpful.
7003 70. exigrep's output now also includes lines that are not associated with any
7004 message, but which match the given pattern. Implemented by a patch from
7005 Martin Sluka, which also tidied up the Perl a bit.
7007 71. Recipient callout verification, like sender verification, was using <> in
7008 the MAIL FROM command. This isn't really the right thing, since the actual
7009 sender may affect whether the remote host accepts the recipient or not. I
7010 have changed it to use the actual sender in the callout; this means that
7011 the cache record is now keyed on a recipient/sender pair, not just the
7012 recipient address. There doesn't seem to be a real danger of callout loops,
7013 since a callout by the remote host to check the sender would use <>.
7014 [SEE ABOVE: changed after hitting problems.]
7016 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
7017 temporary errors. However, in the case of such a code being given after
7018 the end of a data transmission (i.e. after ".") Exim was failing to write
7019 a retry record for the message. (Yes, there was some broken host that was
7020 actually sending 8xx at this point.)
7022 73. An unknown lookup type in a host list could cause Exim to panic-die when
7023 the list was checked. (An example that provoked this was putting <; in the
7024 middle of a list instead of at the start.) If this happened during a DATA
7025 ACL check, a -D file could be left lying around. This kind of configuration
7026 error no longer causes Exim to die; instead it causes a defer error. The
7027 incident is still logged to the main and panic logs.
7029 74. Buglet left over from Exim 3 conversion. The message "too many messages
7030 in one connection" was written to the rejectlog but not the mainlog, except
7031 when address rewriting (yes!) was being logged.
7033 75. Added write_rejectlog option.
7035 76. When a system filter was run not as root (that is, when system_filter_user
7036 was set), the values of the $n variables were not being returned to the
7037 main process; thus, they were not subsequently available in the $sn
7040 77. Added +return_path_on_delivery log selector.
7042 78. A connection timeout was being treated differently from recipients deferred
7043 when testing hosts_max_try with a message that was older than the host's
7044 retry timeout. (The host should not be counted, thus allowing all hosts to
7045 be tried at least once before bouncing.) This may have been the cause of an
7046 occasionally reported bug whereby a message would remain on the queue
7047 longer than the retry timeout, but would be bounced if a delivery was
7048 forced. I say "may" because I never totally pinned down the problem;
7049 setting up timeout/retry tests is difficult. See also the next item.
7051 79. The ultimate address timeout was not being applied to errors that involved
7052 a combination of host plus message (for example, a timeout on a MAIL
7053 command). When an address resolved to a number of possible hosts, and they
7054 were not all tried for each delivery (e.g. because of hosts_max_try), a
7055 message could remain on the queue longer than the retry timeout.
7057 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
7060 81. Fixed an obscure SMTP outgoing bug which required at least the following
7061 conditions: (a) there was another message waiting for the same server;
7062 (b) the server returned 5xx to all RCPT commands in the first message so
7063 that the message was not completed; (c) the server dropped the connection
7064 or gave a negative response to the RSET that Exim sends to abort the
7065 transaction. The observed case was a dropped connection after DATA that had
7066 been sent in pipelining mode. That is, the server had advertised PIPELINING
7067 but was not implementing it correctly. The effect of the bug was incorrect
7068 behaviour, such as trying another host, and this could lead to a crash.
7074 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
7075 and daemon.c were passed as pointers to ints; they should have been
7076 pointers to socklen_t variables (which are typically unsigned ints).
7078 2. Some signed/unsigned type warnings in the os.c file for Linux have been
7081 3. Fixed a really odd bug that affected only the testing scheme; patching a
7082 certain fixed string in the binary changed the value of another string that
7083 happened to be identical to the end of the original first string.
7085 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
7086 name", it returns that address as the IP address. On some operating
7087 systems (e.g. Solaris), it also passes back the IP address string as the
7088 "host name". However, on others (e.g. Linux), it passes back an empty
7089 string. Exim wasn't checking for this, and was changing the host name to an
7090 empty string, assuming it had been canonicalized.
7092 5. Although rare, it is permitted to have more than one PTR record for a given
7093 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
7094 all the names associated with an address, because they do in Solaris.
7095 However, it seems that they do not in Linux for data that comes from the
7096 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
7097 I found this out when I moved to a new Linux workstation and tried to run
7098 the Exim test suite.
7100 To get round this problem I have changed the code so that it now does its
7101 own call to the DNS to look up PTR records when searching for a host name.
7102 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
7103 addresses that are only in /etc/hosts are still found.
7105 This behaviour is, however, controlled by an option called host_lookup_
7106 order, which defaults to "bydns:byaddr". If people want to use the other
7107 order, or indeed, just use one or the other means of lookup, they can
7108 specify it in this variable.
7110 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
7111 some operating systems, this comes back from gethostbyaddr() as an empty
7112 string, and this is what Exim used to test for. However, it seems that in
7113 other systems, "." is yielded. Exim now tests for this case too.
7115 7. The values of check_spool_space and check_log_space are now held internally
7116 as a number of kilobytes instead of an absolute number of bytes. If a
7117 numbers is specified without 'K' or 'M', it is rounded up to the nearest
7118 kilobyte. This means that much larger values can be stored.
7120 8. Exim monitor: an attempt to get the action menu when not actually pointing
7121 at a message produces an empty menu entitled "No message selected". This
7122 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
7123 no entries in it ("Shell widget menu has zero width and/or height"). So I
7124 have added a single, blank menu entry in this case.
7126 9. Added ${quote_local_part.
7128 10. MIME decoding is now applied to the contents of Subject: header lines when
7131 11. Now that a reference to $sender_host_address automatically causes a reverse
7132 lookup to occur if necessary (4.13/18), there is no need to arrange for a
7133 host lookup before query-style lookups in lists that might use this
7134 variable. This has therefore been abolished, and the "net-" prefix is no
7135 longer necessary for query-style lookups.
7137 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
7138 have been a typo for LFLAGS, so it has been changed.
7140 13. The install script calls Exim with "-C /dev/null" in order to find the
7141 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
7142 to be output. However, since Exim outputs its version number before the
7143 error, it didn't break the script. It just looked ugly. I fixed this by
7144 always allowing "-C /dev/null" if the caller is root.
7146 14. Ignore overlarge ACL variable number when reading spool file - insurance
7147 against a later release with more variables having written the file.
7149 15. The standard form for an IPv6 address literal was being rejected by EHLO.
7150 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
7151 this, as well as the form without the "IPv6" on the front.
7153 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
7154 OS/Makefile-Darwin file.
7156 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
7157 only to LDAP libraries that do not have LDAP_OPT_DEREF.
7159 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
7160 variable. However, some OS use ints for inodes. I've added cast to long int
7161 to get rid of the compiler warning.
7163 19. I had forgotten to lock out "/../" in configuration file names when
7164 ALT_CONFIG_PREFIX was set.
7166 20. Routers used for verification do not need to specify transports. However,
7167 if such a router generated a host list, and callout was configured, Exim
7168 crashed, because it could not find a port number from the (non-existent)
7169 transport. It now assumes port 25 in this circumstance.
7171 21. Added the -t option to exigrep.
7173 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
7174 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
7175 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
7176 them is set, LOOKUP_LSEARCH is forced.
7178 23. "exim -bV" now outputs a list of lookups that are included in the binary.
7180 24. Added sender and host information to the "rejected by local_scan()" log
7181 line; previously there was no indication of these.
7183 25. Added .include_if_exists.
7185 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
7186 Linux. It turns out that not all file systems support this. Apparently some
7187 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
7188 do it.) To cope with this, the error EINVAL, which means that sync-ing is
7189 not supported on the file descriptor, is now ignored when Exim is trying to
7190 sync a directory. This applies only to Linux.
7192 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
7194 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
7195 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
7198 29. Added support for maildirsize files from supplied patch (modified a bit).
7200 30. The use of :fail: followed by an empty string could lead Exim to respond to
7201 sender verification failures with (e.g.):
7203 550 Verification failed for <xxx>
7204 550 Sender verify failed
7206 where the first response line was missing the '-' that indicates it is not
7207 the final line of the response.
7209 31. The loop for finding the name of the user that called Exim had a hardwired
7210 limit of 10; it now uses the value of finduser_retries, which is used for
7211 all other user lookups.
7213 32. Added $received_count variable, available in data and not_smtp ACLs, and at
7216 33. Exim was neglecting to zero errno before one call of strtol() when
7217 expanding a string and expecting an integer value. On some systems this
7218 resulted in spurious "integer overflow" errors. Also, it was casting the
7219 result into an int without checking.
7221 34. Testing for a connection timeout using "timeout_connect" in the retry rules
7222 did not work. The code looks as if it has *never* worked, though it appears
7223 to have been documented since at least release 1.62. I have made it work.
7225 35. The "timeout_DNS" error in retry rules, also documented since at least
7226 1.62, also never worked. As it isn't clear exactly what this means, and
7227 clearly it isn't a major issue, I have abolished the feature by treating it
7228 as "timeout", and writing a warning to the main and panic logs.
7230 36. The display of retry rules for -brt wasn't always showing the error code
7233 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
7234 timeout_connect_A, timeout_connect_MX.
7236 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
7237 to the empty sender.
7239 39. The daemon was not analysing the content of -oX till after it had closed
7240 stderr and disconnected from the controlling terminal. This meant that any
7241 syntax errors were only noted on the panic log, and the return code from
7242 the command was 0. By re-arranging the code a little, I've made the
7243 decoding happen first, so such errors now appear on stderr, and the return
7244 code is 1. However, the actual setting up of the sockets still happens in
7245 the disconnected process, so errors there are still only recorded on the
7248 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
7249 connections (as happens on some IP stacks) was logged at start up time as
7250 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
7251 it from "IPv6 and IPv4", which means that two separate sockets are being
7254 41. The debug output for gethostbyname2() or getipnodebyname() failures now
7255 says whether AF_INET or AF_INET6 was passed as an argument.
7257 42. Exiwhat output was messed up when time zones were included in log
7260 43. Exiwhat now gives more information about the daemon's listening ports,
7261 and whether -tls-on-connect was used.
7263 44. The "port" option of the smtp transport is now expanded.
7265 45. A "message" modifier in a "warn" statement in a non-message ACL was being
7266 silently ignored. Now an error message is written to the main and panic
7269 46. There's a new ACL modifier called "logwrite" which writes to a log file
7270 as soon as it is encountered.
7272 47. Added $local_user_uid and $local_user_gid at routing time.
7274 48. Exim crashed when trying to verify a sender address that was being
7277 49. Exim was recognizing only a space character after ".include". It now also
7278 recognizes a tab character.
7280 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
7281 extracting the relevant information from the specification. The man page no
7282 longer contains scrambled data for the -d option, and I've added a section
7283 at the front about calling Exim under different names.
7285 51. Added "extra_headers" argument to the "mail" command in filter files.
7287 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
7290 53. Installed eximstats 1.29.
7292 54. Added transport_filter_timeout as a generic transport option.
7294 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
7295 Cc: header lines. This was required by RFC 822, but it not required by RFC
7298 56. Exim used to add From:, Date:, and Message-Id: header lines to any
7299 incoming messages that did not have them. Now it does so only if the
7300 message originates locally, that is, if there is no associated remote host
7301 address. When Resent- header lines are present, this applies to the Resent-
7302 lines rather than the non-Resent- lines.
7304 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
7305 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
7307 58. Messages for configuration errors now include the name of the main
7308 configuration file - useful now that there may be more than one file in a
7309 list (.included file names were always shown).
7311 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
7312 for those rare installations that do not start the daemon as root or run it
7313 setuid root. I've cut out the call to initgroups() if the daemon is not
7316 60. The Exim user and group can now be bound into the binary as text strings
7317 that are looked up at the start of Exim's processing.
7319 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
7321 62. Added $mailstore_basename variable.
7323 63. Installed patch to sieve.c from Michael Haardt.
7325 64. When Exim failed to open the panic log after failing to open the main log,
7326 the original message it was trying to log was written to stderr and debug
7327 output, but if they were not available (the usual case in production), it
7328 was lost. Now it is written to syslog before the two lines that record the
7329 failures to open the logs.
7331 65. Users' Exim filters run in subprocesses under the user's uid. It is
7332 possible for a "deliver" command or an alias in a "personal" command to
7333 provoke an address rewrite. If logging of address rewriting is configured,
7334 this fails because the process is not running as root or exim. There may be
7335 a better way of dealing with this, but for the moment (because 4.30 needs
7336 to be released), I have disabled address rewrite logging when running a
7337 filter in a non-root, non-exim process.
7343 1. The buildconfig auxiliary program wasn't quoting the value set for
7344 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
7345 not defined. This bug was masked in 4.22 by the effect that was fixed in
7348 2. Some messages that were rejected after a message id was allocated were
7349 shown as "incomplete" by exigrep. It no longer does this for messages that
7350 are rejected by local_scan() or the DATA or non-SMTP ACLs.
7352 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
7353 have allow_domain_literals set, the ID did not get logged in the <= line.
7354 Domain literals are now always recognized in Message-ID: header lines.
7356 4. The first argument for a ${extract expansion item is the key name or field
7357 number. Leading and trailing spaces in this item were not being ignored,
7358 causing some misleading effects.
7360 5. When deliver_drop_privilege was set, single queue runner processes started
7361 manually (i.e. by the command "exim -q") or by the daemon (which uses the
7362 same command in the process it spins off) were not dropping privilege.
7364 6. When the daemon running as "exim" started a queue runner, it always
7365 re-executed Exim in the spun-off process. This is a waste of effort when
7366 deliver_drop_privilege is set. The new process now just calls the
7367 queue-runner function directly.
7373 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
7376 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
7377 ignored. Though the use of -oP was forcing the writing of a pid file, it
7378 was always written to the default place.
7380 3. If the message "no IP address found for host xxxx" is generated during
7381 incoming verification, it is now followed by identification of the incoming
7382 connection (so you can more easily find what provoked it).
7384 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
7386 5. Added some features to "harden" Exim a bit more against certain attacks:
7388 (a) There is now a build-time option called FIXED_NEVER_USERS that can
7389 be put in Local/Makefile. This is like the never_users runtime option,
7390 but it cannot be overridden. The default setting is "root".
7392 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
7393 prefix string with which any file named in a -C command line option
7396 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
7397 is retained for -C and -D only if the caller of Exim is root. Without
7398 it, the exim user may also use -C and -D and retain privilege.
7400 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
7401 command line option is disabled.
7403 6. Macro names set by the -D option must start with an upper case letter, just
7404 like macro names defined in the configuration file.
7406 7. Added "dereference=" facility to LDAP.
7408 8. Two instances of the typo "uknown" in the source files are fixed.
7410 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
7411 the Configure-Makefile script screwed up while processing it.
7413 10. Incorporated PCRE 4.4.
7415 11. The SMTP synchronization check was not operating right at the start of an
7416 SMTP session. For example, it could not catch a HELO sent before the client
7417 waited for the greeting. There is now a check for outstanding input at the
7418 point when the greeting is written. Because of the duplex, asynchronous
7419 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
7420 way, but not yet received, when the check is performed.
7422 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
7423 on TCP/IP sockets, because this apparently causes some broken clients to
7426 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
7427 unchanged) from the Cygwin maintainer.
7429 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
7430 was supported instead of testing for mbx. Effectively a typo.
7432 15. The spa authenticator server code was not checking that the input it
7433 received was valid base64.
7435 16. The debug output line for the "set" modifier in ACLs was not showing the
7436 name of the variable that was being set.
7438 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
7440 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
7441 lookup on its own. Something else was needed to trigger the lookup. For
7442 example, a match in host_lookup or the need for a host name in a host list.
7443 Now, if $sender_host_name is referenced and the host name has not yet been
7444 looked up, a lookup is performed. If the lookup fails, the variable remains
7445 empty, and $host_lookup_failed is set to "1".
7447 19. Added "eqi" as a case-independent comparison operator.
7449 20. The saslauthd authentication condition could segfault if neither service
7450 nor realm was specified.
7452 21. If an overflowing value such as "2048M" was set for message_size_limit, the
7453 error message that was logged was misleading, and incoming SMTP
7454 connections were dropped. The message is now more accurate, and temporary
7455 errors are given to SMTP connections.
7457 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
7458 (except RSET) with a 421 error, until QUIT is received. However, it was
7459 failing to send a response to QUIT.
7461 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
7462 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
7463 ACL is now run after the helo_try_verify_hosts code.
7465 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
7466 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
7467 case-independent, so other case variants are also recognized). Apparently
7468 some systems use these upper case variants.
7470 25. If more than two messages were waiting for the same host, and a transport
7471 filter was specified for the transport, Exim sent two messages over the
7472 same TCP/IP connection, and then failed with "socket operation on non-
7473 socket" when it tried to send the third.
7475 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
7477 27. The extern definition of crypt16() in expand.c was not being excluded when
7478 the OS had its own crypt16() function.
7480 28. Added bounce_return_body as a new option, and bounce_return_size_limit
7481 as a preferred synonym for return_size_limit, both as an option and as an
7484 29. Added LIBS=-liconv to OS/Makefile-OSF1.
7486 30. Changed the default configuration ACL to relax the local part checking rule
7487 for addresses that are not in any local domains. For these addresses,
7488 slashes and pipe symbols are allowed within local parts, but the sequence
7489 /../ is explicitly forbidden.
7491 31. SPA server authentication was not clearing the challenge buffer before
7494 32. log_message in a "warn" ACL statement was writing to the reject log as
7495 well as to the main log, which contradicts the documentation and doesn't
7496 seem right (because no rejection is happening). So I have stopped it.
7498 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
7499 However, I am unable to do any testing of this.
7501 34. Fixed an infelicity in the appendfile transport. When checking directories
7502 for a mailbox, to see if any needed to be created, it was accidentally
7503 using path names with one or more superfluous leading slashes; tracing
7504 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
7506 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
7507 contents are added to the log line that is written for every discarded
7508 recipient. (Previously a log_message setting was ignored.)
7510 36. The ${quote: operator now quotes the string if it is empty.
7512 37. The install script runs exim in order to find its version number. If for
7513 some reason other than non-existence or emptiness, which it checks, it
7514 could not run './exim', it was installing it with an empty version number,
7515 i.e. as "exim-". This error state is now caught, and the installation is
7518 38. An argument was missing from the function that creates an error message
7519 when Exim fails to connect to the socket for saslauthd authentication.
7520 This could cause Exim to crash, or give a corrupted message.
7522 39. Added isip, isip4, and isip6 to ${if conditions.
7524 40. The ACL variables $acl_xx are now saved with the message, and can be
7525 accessed later in routers, transports, and filters.
7527 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
7528 strings in the file are not string-expanded.
7530 42. If a MAIL command specified a SIZE value that was too large to fit into an
7531 int variable, the check against message_size_limit failed. Such values are
7532 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
7533 day this will have to be increased, but I don't think I want to be around
7534 when emails are that large.
7541 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
7542 iconv() is not standard in FreeBSD.
7544 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
7545 IPv6 enabled. The observed symptom was a segmentation fault on return from
7546 the function os_common_find_running_interfaces() in src/os.c.
7548 3. In the check_special_case() function in daemon.c I had used "errno" as an
7549 argument name, which causes warnings on some systems. This was basically a
7550 typo, since it was named "eno" in the comments!
7552 4. The code that waits for the clock to tick (at a resolution of some fraction
7553 of a second) so as to ensure message-id uniqueness was always waiting for
7554 at least one whole tick, when it could have waited for less. [This is
7555 almost certainly not relevant at current processor speeds, where it is
7556 unlikely to ever wait at all. But we try to future-proof.]
7558 5. The function that sleeps for a time interval that includes fractions of a
7559 second contained a race. It did not block SIGALRM between setting the
7560 timer, and suspending (a couple of lines later). If the interval was short
7561 and the sigsuspend() was delayed until after it had expired, the suspension
7562 never ended. On busy systems this could lead to processes getting stuck for
7565 6. Some uncommon configurations may cause a lookup to happen in a queue runner
7566 process, before it forks any delivery processes. The open lookup caching
7567 mechanism meant that the open file or database connection was passed into
7568 the delivery process. The problem was that delivery processes always tidy
7569 up cached lookup data. This could cause a problem for the next delivery
7570 process started by the queue runner, because the external queue runner
7571 process does not know about the closure. So the next delivery process
7572 still has data in the lookup cache. In the case of a file lookup, there was
7573 no problem because closing a file descriptor in a subprocess doesn't affect
7574 the parent. However, if the lookup was caching a connection to a database,
7575 the connection was closed, and the second delivery process was likely to
7576 see errors such as "PGSQL: query failed: server closed the connection
7577 unexpectedly". The problem has been fixed by closing all cached lookups
7578 in a queue runner before running a delivery process.
7580 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
7581 seem to have the "const" qualifier which it has on other OS. I've
7584 8. Change 4.21/2 was too strict. It is only if there are two authenticators
7585 *of the same type* (client or server) with the same public name that an
7586 error should be diagnosed.
7588 9. When Exim looked up a host name for an IP address, but failed to find the
7589 original IP address when looking up the host name (a safety check), it
7590 output the message "<ip address> does not match any IP for NULL", which was
7591 confusing, to say the least. The bug was that the host name should have
7592 appeared instead of "NULL".
7594 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
7595 user that is built into the binary, and the -C or -D options is used, root
7596 privilege is dropped before the configuration file is read. In addition,
7597 logging is switched to stderr instead of the normal log files. If the
7598 configuration then re-defines the Exim user, the unprivileged environment
7599 is probably not what is expected, so Exim logs a panic warning message (but
7602 However, if deliver_drop_privilege is set, the unprivileged state may well
7603 be exactly what is intended, so the warning has been cut out in that case,
7604 and Exim is allowed to try to write to its normal log files.
7610 1. smtp_return_error_details was not giving details for temporary sender
7611 or receiver verification errors.
7613 2. Diagnose a configuration error if two authenticators have the same public
7616 3. Exim used not to create the message log file for a message until the first
7617 delivery attempt. This could be confusing when incoming messages were held
7618 for policy or load reasons. The message log file is now created at the time
7619 the message is received, and an initial "Received" line is written to it.
7621 4. The automatically generated man page for command line options had a minor
7622 bug that caused no ill effects; however, a more serious problem was that
7623 the procedure for building the man page automatically didn't always
7624 operate. Consequently, release 4.20 contains an out-of-date version. This
7625 shouldn't happen again.
7627 5. When building Exim with embedded Perl support, the script that builds the
7628 Makefile was calling 'perl' to find its compile-time parameters, ignoring
7629 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
7631 6. The freeze_tell option was not being used for messages that were frozen on
7632 arrival, either by an ACL or by local_scan().
7634 7. Added the smtp_incomplete_transaction log selector.
7636 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
7637 was accepting AUTH without a new EHLO.
7639 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
7640 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
7643 10. Logging of TCP/IP connections (when configured) now happens in the main
7644 daemon process instead of the child process, so that the TCP/IP connection
7645 count is more accurate (but it can never be perfect).
7647 11. The use of "drop" in a nested ACL was not being handled correctly in the
7648 outer ACL. Now, if condition failure induced by the nested "drop" causes
7649 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
7650 or "require"), the connection is dropped.
7652 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
7653 that yield "discard" can now be used with an "accept" or a "discard" verb,
7654 but an error is generated for any others (because I can't see a useful way
7655 to define what should happen).
7657 13. When an ACL is read dynamically from a file (or anywhere else), the lines
7658 are now processed in the same way as lines in the Exim configuration file.
7659 In particular, continuation lines are supported.
7661 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
7663 15. Added -ti meaning -t -i.
7665 16. Check for letters, digits, hyphens, and dots in the names of dnslist
7666 domains, and warn by logging if others are found.
7668 17. At least on BSD, alignment is not guaranteed for the array of ifreq's
7669 returned from GIFCONF when Exim is trying to find the list of interfaces on
7670 a host. The code in os.c has been modified to copy each ifreq to an aligned
7671 structure in all cases.
7673 Also, in some cases, the returned ifreq's were being copied to a 'struct
7674 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
7675 means the last couple of bytes of an IPv6 address could be chopped if the
7676 ifreq contained only a normal sockaddr (14 bytes storage).
7678 18. Named domain lists were not supported in the hosts_treat_as_local option.
7679 An entry such as +xxxx was not recognized, and was treated as a literal
7682 19. Ensure that header lines added by a DATA ACL are included in the reject log
7683 if the ACL subsequently rejects the message.
7685 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
7686 MD5 (which is deprecated).
7688 21. When testing a filter file using -bf, Exim was writing a message when it
7689 took the sender from a "From " line in the message, but it was not doing so
7690 when it took $return_path from a Return-Path: header line. It now does.
7692 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
7693 with a valid header line field name (a series of printing characters
7694 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
7696 23. Changed "disc" in the source to "disk" to conform to the documentation and
7697 the book and for uniformity.
7699 24. Ignore Sendmail's -Ooption=value command line item.
7701 25. When execve() failed while trying to run a command in a pipe transport,
7702 Exim was returning EX_UNAVAILABLE (69) from the subprocess. However, this
7703 could be confused with a return value of 69 from the command itself. This
7704 has been changed to 127, the value the shell returns if it is asked to run
7705 a non-existent command. The wording for the related log line suggests a
7706 non-existent command as the problem.
7708 26. If received_header_text expands to an empty string, do not add a Received:
7709 header line to the message. (Well, it adds a token one on the spool, but
7710 marks it "old" so that it doesn't get used or transmitted.)
7712 27. Installed eximstats 1.28 (addition of -nt option).
7714 28. There was no check for failure on the call to getsockname() in the daemon
7715 code. This can fail if there is a shortage of resources on the system, with
7716 ENOMEM, for example. A temporary error is now given on failure.
7718 29. Contrary to the C standard, it seems that in some environments, the
7719 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
7720 program. Exim now does this explicitly; it affects the formatting of
7721 timestamps using strftime().
7723 30. If exiqsumm was given junk data, it threw up some uninitialized variable
7724 complaints. I've now initialized all the variables, to avoid this.
7726 32. Header lines added by a system filter were not being "seen" during
7727 transport-time rewrites.
7729 33. The info_callback() function passed to OpenSSL is set up with type void
7730 (*)(SSL *, int, int), as described somewhere. However, when calling the
7731 function (actually a macro) that sets it up, the type void(*)() is
7732 expected. I've put in a cast to prevent warnings from picky compilers.
7734 34. If a DNS black list lookup found a CNAME record, but there were no A
7735 records associated with the domain it pointed at, Exim crashed.
7737 35. If a DNS black list lookup returned more than one A record, Exim ignored
7738 all but the first. It now scans all returned addresses if a particular IP
7739 value is being sought. In this situation, the contents of the
7740 $dnslist_value variable are a list of all the addresses, separated by a
7743 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
7744 to accept a host name and all its aliases if the forward lookup for any of
7745 them yielded the IP address of the incoming connection. Now it accepts only
7746 those names whose forward lookup yields the correct IP address. Any other
7747 names are discarded. This closes a loophole whereby a rogue DNS
7748 administrator could create reverse DNS records to break through a
7749 wildcarded host restriction in an ACL.
7751 37. If a user filter or a system filter that ran in a subprocess used any of
7752 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
7753 the wrong values were passed to the pipe command ($thisaddress had the
7754 value of $0, $0 had the value of $1, etc). This bug was introduced by
7755 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
7757 38. Improved the line breaking for long SMTP error messages from ACLs.
7758 Previously, if there was no break point between 40 and 75 characters, Exim
7759 left the rest of the message alone. Two changes have been made: (a) I've
7760 reduced the minimum length to 35 characters; (b) if it can't find a break
7761 point between 35 and 75 characters, it looks ahead and uses the first one
7762 that it finds. This may give the occasional overlong line, but at least the
7763 remaining text gets split now.
7765 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
7766 file descriptors might be low, and that setting 1000 would always raise it.
7767 It turns out that in some environments, the limit is already over 1000 and
7768 that lowering it causes trouble. So now Exim takes care not to decrease it.
7770 40. When delivering a message, the value of $return_path is set to $sender_
7771 address at the start of routing (routers may change the value). By an
7772 oversight, this default was not being set up when an address was tested by
7773 -bt or -bv, which affected the outcome if any router or filter referred to
7776 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
7777 log only when "message" or "log_message" are set. However, if one of the
7778 conditions was an address verification, or a call to a nested ACL, the
7779 messages generated by the underlying test were being passed through. This
7780 no longer happens. The underlying message is available in $acl_verify_
7781 message for both "message" and "log_message" expansions, so it can be
7782 passed through if needed.
7784 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
7785 new expansion $bh_ to give the encoded byte string without charset
7786 translation. Translation happens only if iconv() is available; HAVE_ICONV
7787 indicates this at build time. HEADERS_CHARSET gives the charset to
7788 translate to; headers_charset can change it in the configuration, and
7789 "headers charset" can change it in an individual filter file.
7791 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
7792 that creates RFC 2047 encoded "words" labels them as that charset instead
7793 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
7794 expansion operator; (ii) when Exim creates a From: line for a local
7795 message; (iii) when a header line is rewritten to include a "phrase" part.
7797 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
7798 buggy, causing it to skip the first lines of messages whose message ID
7799 ended in 'D'. This would not have bitten before Exim release 4.14, because
7800 message IDs were unlikely to end in 'D' before then. The effect was to have
7801 incorrect size information for certain domains.
7803 45. #include "config.h" was missing at the start of the crypt16.c module. This
7804 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
7807 46. If there was a timeout during a "random" callout check, Exim treated it as
7808 a failure of the random address, and carried on sending RSET and the real
7809 address. If the delay was just some slowness somewhere, the response to the
7810 original RCPT would be taken as a response to RSET and so on, causing
7811 mayhem of various kinds.
7813 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
7814 when I implemented it. It didn't allow for the fact that some option values
7815 may legitimately be negative (e.g. size_addition), and it didn't even do
7816 the right test for positive values.
7818 48. Domain names in DNS records are case-independent. Exim always looks them up
7819 in lower case. Some resolvers return domain names in exactly the case they
7820 appear in the zone file, that is, they may contain uppercase letters. Not
7821 all resolvers do this - some return always lower case. Exim was treating a
7822 change of case by a resolver as a change of domain, similar to a widening
7823 of a domain abbreviation. This triggered its re-routing code and so it was
7824 trying to route what was effectively the same domain again. This normally
7825 caused routing to fail (because the router wouldn't handle the domain
7826 twice). Now Exim checks for this case specially, and just changes the
7827 casing of the domain that it ultimately uses when it transmits the message
7830 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
7833 50. If a filter generated a file delivery with a non-absolute name (possible if
7834 no home directory exists for the router), the forbid_file option was not
7837 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
7838 the existing equality matching.
7840 52. Exim was using ints instead of ino_t variables in some places where it was
7841 dealing with inode numbers.
7843 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
7844 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
7845 called TMPDIR, and if it finds it is different, it changes its value.
7847 54. The smtp_printf() function is now made available to local_scan() so
7848 additional output lines can be written before returning. There is also an
7849 smtp_fflush() function to enable the detection of a dropped connection.
7850 The variables smtp_input and smtp_batched_input are exported to
7853 55. Changed the default runtime configuration: the message "Unknown user"
7854 has been removed from the ACL, and instead placed on the localuser router,
7855 using the cannot_route_message feature. This means that any verification
7856 failures that generate their own messages won't get overridden. Similarly,
7857 the "Unrouteable address" message that was in the ACL for unverifiable
7858 relay addresses has also been removed.
7860 56. Added hosts_avoid_esmtp to the smtp transport.
7862 57. The exicyclog script was not checking for the esoteric option
7863 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
7864 will work only if exicyclog is run under the appropriate euid.
7866 58. Following a discussion on the list, the rules by which Exim recognises line
7867 endings on incoming messages have been changed. The -dropcr and drop_cr
7868 options are now no-ops, retained only for backwards compatibility. The
7869 following line terminators are recognized: LF CRLF CR. However, special
7870 processing applies to CR:
7872 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
7873 nor a local message in the state where . is a terminator.
7875 (ii) If a bare CR is encountered in a header line, an extra space is added
7876 after the line terminator so as not to end the header. The reasoning
7877 behind this is that bare CRs in header lines are most likely either
7878 to be mistakes, or people trying to play silly games.
7880 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
7881 was being incorrectly given as 18 bytes larger than it should have been.
7882 This is a VOB (very old bug).
7884 60. This may never have affected anything current, but just in case it has:
7885 When the local host is found other than at the start of a list of hosts,
7886 the local host, those with the same MX, and any that follow, are discarded.
7887 When the list in question was part of a longer list of hosts, the following
7888 hosts (not currently being processed) were also being discarded. This no
7889 longer happens. I'm not sure if this situation could ever has previously
7892 61. Added the "/MX" feature to lists of hosts in the manualroute and query
7895 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
7896 header. This is something that is recommended in a new Internet Draft, and
7897 is something that is documented as being done by Sendmail. There are two
7898 possible values. For messages generated by the autoreply transport, Exim
7901 Auto-Submitted: auto-replied
7903 whereas for all other generated messages (e.g. bounces) it adds
7905 Auto-Submitted: auto-generated
7907 63. The "personal" condition in filters now includes a test for the
7908 Auto-Submitted: header. If it contains the string "auto-" the message it
7909 not considered personal.
7911 64. Added rcpt_include_affixes as a generic transport option.
7913 65. Added queue_only_override (default true).
7915 66. Added the syslog_duplication option.
7917 67. If what should have been the first header line of a message consisted of
7918 a space followed by a colon, Exim was mis-interpreting it as a header line.
7919 It isn't of course - it is syntactically invalid and should therefore be
7920 treated as the start of the message body. The misbehaviour could have
7921 caused a number of strange effects, including loss of data in subsequent
7922 header lines, and spool format errors.
7924 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
7925 client host had authenticated. This control can now be exercised by an ACL
7926 for more flexibility.
7928 69. By default, callouts do not happen when testing with -bh. There is now a
7929 variant, -bhc, which does actually run the callout code, including
7930 consulting and updating the callout cache.
7932 70. Added support for saslauthd authentication, courtesy of Alexander
7935 71. If statvfs() failed on the spool or log directories while checking their
7936 size for availability, Exim confusingly gave the error "space shortage".
7937 Furthermore, in debugging mode it crashed with a floating point exception.
7938 These checks are done if check_{spool,log}_{space,inodes} are set, and when
7939 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
7940 serious problem, Exim now writes to the main and panic logs when this
7941 happens, with details of the failure. It then refuses to accept the
7942 incoming message, giving the message "spool directory problem" or "log
7943 directory problem" with a 421 code for SMTP messages.
7945 72. When Exim is about to re-exec itself, it ensures that the file descriptors
7946 0, 1, and 2 exist, because some OS complain for execs without them (see
7947 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
7948 descriptors. However, the code omitted to check that the open succeeded,
7949 causing mysterious errors if for some reason the permissions on /dev/null
7950 got screwed. Now Exim writes a message to the main and panic logs, and
7951 bombs out if it can't open /dev/null.
7953 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
7954 interact so that it is all more flexible. It is supposed to remain
7955 backwards compatible. Also added extra_local_interfaces.
7957 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
7958 to bomb out with an assertion failure - to the client this appears as a
7959 connection drop. This problem occurs in the part of the code that was taken
7960 from the Samba project. Fortunately, the assertion is in a very simple
7961 function, so I have fixed this by reproducing the function inline in the
7962 one place where it is called, and arranging for authentication to fail
7963 instead of killing the process with assert().
7965 75. The SPA client code was not working when the server requested OEM rather
7966 than Unicode encoding.
7968 76. Added code to make require_files with a specific uid setting more usable in
7969 the case where statting the file as root fails - usually a non-root-mounted
7970 NFS file system. When this happens and the failure is EACCES, Exim now
7971 forks a subprocess and does the per-uid checking as the relevant uid.
7973 77. Added process_log_path.
7975 78. If log_file_path was not explicitly set, a setting of check_log_space or
7976 check_log_inodes was ignored.
7978 79. If a space check for the spool or log partitions fails, the incident is now
7979 logged. Of course, in the latter case the data may get lost...
7981 80. Added the %p formatting code to string_format() so that it can be used to
7982 print addresses in debug_print(). Adjusted all the address printing in the
7983 debugging in store.c to use %p rather than %d.
7985 81. There was a concern that a line of code in smtp_in.c could overflow a
7986 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
7987 initially expressed, the concern was not well-founded, because trailing
7988 spaces are removed early. However, if the trailing spaces were followed by
7989 a NULL, they did not get removed, so the overflow was possible. Two fixes
7992 (a) I re-wrote the offending code in a cleaner fashion.
7993 (b) If an incoming SMTP command contains a NULL character, it is rejected
7996 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
7997 runs initgroups(), so that if the Exim user is in any additional groups,
7998 they will be used during message reception.
8004 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob